National Academies Press: OpenBook

Alternative Aircraft Anti-Icing Formulations with Reduced Aquatic Toxicity and Biochemical Oxygen Demand (2010)

Chapter: Section 2: Summary of Candidate Deicer Components Evaluated in Test Program

« Previous: Section 1: Introduction
Page 7
Suggested Citation:"Section 2: Summary of Candidate Deicer Components Evaluated in Test Program." National Academies of Sciences, Engineering, and Medicine. 2010. Alternative Aircraft Anti-Icing Formulations with Reduced Aquatic Toxicity and Biochemical Oxygen Demand. Washington, DC: The National Academies Press. doi: 10.17226/22962.
×
Page 7
Page 8
Suggested Citation:"Section 2: Summary of Candidate Deicer Components Evaluated in Test Program." National Academies of Sciences, Engineering, and Medicine. 2010. Alternative Aircraft Anti-Icing Formulations with Reduced Aquatic Toxicity and Biochemical Oxygen Demand. Washington, DC: The National Academies Press. doi: 10.17226/22962.
×
Page 8
Page 9
Suggested Citation:"Section 2: Summary of Candidate Deicer Components Evaluated in Test Program." National Academies of Sciences, Engineering, and Medicine. 2010. Alternative Aircraft Anti-Icing Formulations with Reduced Aquatic Toxicity and Biochemical Oxygen Demand. Washington, DC: The National Academies Press. doi: 10.17226/22962.
×
Page 9
Page 10
Suggested Citation:"Section 2: Summary of Candidate Deicer Components Evaluated in Test Program." National Academies of Sciences, Engineering, and Medicine. 2010. Alternative Aircraft Anti-Icing Formulations with Reduced Aquatic Toxicity and Biochemical Oxygen Demand. Washington, DC: The National Academies Press. doi: 10.17226/22962.
×
Page 10
Page 11
Suggested Citation:"Section 2: Summary of Candidate Deicer Components Evaluated in Test Program." National Academies of Sciences, Engineering, and Medicine. 2010. Alternative Aircraft Anti-Icing Formulations with Reduced Aquatic Toxicity and Biochemical Oxygen Demand. Washington, DC: The National Academies Press. doi: 10.17226/22962.
×
Page 11
Page 12
Suggested Citation:"Section 2: Summary of Candidate Deicer Components Evaluated in Test Program." National Academies of Sciences, Engineering, and Medicine. 2010. Alternative Aircraft Anti-Icing Formulations with Reduced Aquatic Toxicity and Biochemical Oxygen Demand. Washington, DC: The National Academies Press. doi: 10.17226/22962.
×
Page 12
Page 13
Suggested Citation:"Section 2: Summary of Candidate Deicer Components Evaluated in Test Program." National Academies of Sciences, Engineering, and Medicine. 2010. Alternative Aircraft Anti-Icing Formulations with Reduced Aquatic Toxicity and Biochemical Oxygen Demand. Washington, DC: The National Academies Press. doi: 10.17226/22962.
×
Page 13
Page 14
Suggested Citation:"Section 2: Summary of Candidate Deicer Components Evaluated in Test Program." National Academies of Sciences, Engineering, and Medicine. 2010. Alternative Aircraft Anti-Icing Formulations with Reduced Aquatic Toxicity and Biochemical Oxygen Demand. Washington, DC: The National Academies Press. doi: 10.17226/22962.
×
Page 14
Page 15
Suggested Citation:"Section 2: Summary of Candidate Deicer Components Evaluated in Test Program." National Academies of Sciences, Engineering, and Medicine. 2010. Alternative Aircraft Anti-Icing Formulations with Reduced Aquatic Toxicity and Biochemical Oxygen Demand. Washington, DC: The National Academies Press. doi: 10.17226/22962.
×
Page 15
Page 16
Suggested Citation:"Section 2: Summary of Candidate Deicer Components Evaluated in Test Program." National Academies of Sciences, Engineering, and Medicine. 2010. Alternative Aircraft Anti-Icing Formulations with Reduced Aquatic Toxicity and Biochemical Oxygen Demand. Washington, DC: The National Academies Press. doi: 10.17226/22962.
×
Page 16
Page 17
Suggested Citation:"Section 2: Summary of Candidate Deicer Components Evaluated in Test Program." National Academies of Sciences, Engineering, and Medicine. 2010. Alternative Aircraft Anti-Icing Formulations with Reduced Aquatic Toxicity and Biochemical Oxygen Demand. Washington, DC: The National Academies Press. doi: 10.17226/22962.
×
Page 17
Page 18
Suggested Citation:"Section 2: Summary of Candidate Deicer Components Evaluated in Test Program." National Academies of Sciences, Engineering, and Medicine. 2010. Alternative Aircraft Anti-Icing Formulations with Reduced Aquatic Toxicity and Biochemical Oxygen Demand. Washington, DC: The National Academies Press. doi: 10.17226/22962.
×
Page 18
Page 19
Suggested Citation:"Section 2: Summary of Candidate Deicer Components Evaluated in Test Program." National Academies of Sciences, Engineering, and Medicine. 2010. Alternative Aircraft Anti-Icing Formulations with Reduced Aquatic Toxicity and Biochemical Oxygen Demand. Washington, DC: The National Academies Press. doi: 10.17226/22962.
×
Page 19
Page 20
Suggested Citation:"Section 2: Summary of Candidate Deicer Components Evaluated in Test Program." National Academies of Sciences, Engineering, and Medicine. 2010. Alternative Aircraft Anti-Icing Formulations with Reduced Aquatic Toxicity and Biochemical Oxygen Demand. Washington, DC: The National Academies Press. doi: 10.17226/22962.
×
Page 20
Page 21
Suggested Citation:"Section 2: Summary of Candidate Deicer Components Evaluated in Test Program." National Academies of Sciences, Engineering, and Medicine. 2010. Alternative Aircraft Anti-Icing Formulations with Reduced Aquatic Toxicity and Biochemical Oxygen Demand. Washington, DC: The National Academies Press. doi: 10.17226/22962.
×
Page 21
Page 22
Suggested Citation:"Section 2: Summary of Candidate Deicer Components Evaluated in Test Program." National Academies of Sciences, Engineering, and Medicine. 2010. Alternative Aircraft Anti-Icing Formulations with Reduced Aquatic Toxicity and Biochemical Oxygen Demand. Washington, DC: The National Academies Press. doi: 10.17226/22962.
×
Page 22
Page 23
Suggested Citation:"Section 2: Summary of Candidate Deicer Components Evaluated in Test Program." National Academies of Sciences, Engineering, and Medicine. 2010. Alternative Aircraft Anti-Icing Formulations with Reduced Aquatic Toxicity and Biochemical Oxygen Demand. Washington, DC: The National Academies Press. doi: 10.17226/22962.
×
Page 23
Page 24
Suggested Citation:"Section 2: Summary of Candidate Deicer Components Evaluated in Test Program." National Academies of Sciences, Engineering, and Medicine. 2010. Alternative Aircraft Anti-Icing Formulations with Reduced Aquatic Toxicity and Biochemical Oxygen Demand. Washington, DC: The National Academies Press. doi: 10.17226/22962.
×
Page 24
Page 25
Suggested Citation:"Section 2: Summary of Candidate Deicer Components Evaluated in Test Program." National Academies of Sciences, Engineering, and Medicine. 2010. Alternative Aircraft Anti-Icing Formulations with Reduced Aquatic Toxicity and Biochemical Oxygen Demand. Washington, DC: The National Academies Press. doi: 10.17226/22962.
×
Page 25
Page 26
Suggested Citation:"Section 2: Summary of Candidate Deicer Components Evaluated in Test Program." National Academies of Sciences, Engineering, and Medicine. 2010. Alternative Aircraft Anti-Icing Formulations with Reduced Aquatic Toxicity and Biochemical Oxygen Demand. Washington, DC: The National Academies Press. doi: 10.17226/22962.
×
Page 26
Page 27
Suggested Citation:"Section 2: Summary of Candidate Deicer Components Evaluated in Test Program." National Academies of Sciences, Engineering, and Medicine. 2010. Alternative Aircraft Anti-Icing Formulations with Reduced Aquatic Toxicity and Biochemical Oxygen Demand. Washington, DC: The National Academies Press. doi: 10.17226/22962.
×
Page 27
Page 28
Suggested Citation:"Section 2: Summary of Candidate Deicer Components Evaluated in Test Program." National Academies of Sciences, Engineering, and Medicine. 2010. Alternative Aircraft Anti-Icing Formulations with Reduced Aquatic Toxicity and Biochemical Oxygen Demand. Washington, DC: The National Academies Press. doi: 10.17226/22962.
×
Page 28
Page 29
Suggested Citation:"Section 2: Summary of Candidate Deicer Components Evaluated in Test Program." National Academies of Sciences, Engineering, and Medicine. 2010. Alternative Aircraft Anti-Icing Formulations with Reduced Aquatic Toxicity and Biochemical Oxygen Demand. Washington, DC: The National Academies Press. doi: 10.17226/22962.
×
Page 29
Page 30
Suggested Citation:"Section 2: Summary of Candidate Deicer Components Evaluated in Test Program." National Academies of Sciences, Engineering, and Medicine. 2010. Alternative Aircraft Anti-Icing Formulations with Reduced Aquatic Toxicity and Biochemical Oxygen Demand. Washington, DC: The National Academies Press. doi: 10.17226/22962.
×
Page 30
Page 31
Suggested Citation:"Section 2: Summary of Candidate Deicer Components Evaluated in Test Program." National Academies of Sciences, Engineering, and Medicine. 2010. Alternative Aircraft Anti-Icing Formulations with Reduced Aquatic Toxicity and Biochemical Oxygen Demand. Washington, DC: The National Academies Press. doi: 10.17226/22962.
×
Page 31
Page 32
Suggested Citation:"Section 2: Summary of Candidate Deicer Components Evaluated in Test Program." National Academies of Sciences, Engineering, and Medicine. 2010. Alternative Aircraft Anti-Icing Formulations with Reduced Aquatic Toxicity and Biochemical Oxygen Demand. Washington, DC: The National Academies Press. doi: 10.17226/22962.
×
Page 32
Page 33
Suggested Citation:"Section 2: Summary of Candidate Deicer Components Evaluated in Test Program." National Academies of Sciences, Engineering, and Medicine. 2010. Alternative Aircraft Anti-Icing Formulations with Reduced Aquatic Toxicity and Biochemical Oxygen Demand. Washington, DC: The National Academies Press. doi: 10.17226/22962.
×
Page 33
Page 34
Suggested Citation:"Section 2: Summary of Candidate Deicer Components Evaluated in Test Program." National Academies of Sciences, Engineering, and Medicine. 2010. Alternative Aircraft Anti-Icing Formulations with Reduced Aquatic Toxicity and Biochemical Oxygen Demand. Washington, DC: The National Academies Press. doi: 10.17226/22962.
×
Page 34

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.

2-1 SECTION 2 SUMMARY OF CANDIDATE DEICER COMPONENTS EVALUATED IN TEST PROGRAM The initial phase of the research focused on identifying alternatives for four major deicer components based on their contributions to BOD and toxicity: • FPDs: 26 candidates • Thickeners: 5 candidates • Surfactants: 19 candidates • Corrosion Inhibitors: 17 candidates Also identified were two candidate anti-caking additives for use in sodium formate runway deicers and one candidate antifoam for use in deicing formulations. A combination of molecular modeling, database searches, and literature searches were used to identify candidates for each of these component classes. The selection approach and recommended candidates are detailed in the following subsections. FPD Candidates for ADFs and AAFs The following chemicals were identified in the literature as either current or proposed FPDs in ADFs or AAFs: • 1,2-Propylene glycol • Glycerol • 1,3-Butanediol • Mannitol • 1,3-Propylene glycol • Pentaerythritol • Diethylene glycol • Sorbitol • Dipropylene glycol • Triethylene glycol • Erythritol • Urea • Ethyl lactate • Xylitol • Ethylene glycol The general characteristics of viable FPDs were identified by examining this list and considering some of the fundamental relationships between chemical structure and physical properties. These characteristics include: • FPDs have low molecular weight. The largest molecule contains six carbon atoms. This makes sense because freezing point depression is related to molar concentration. • FPDs exhibit strong hydrogen bonding because of the presence of alcohol or amine groups. The presence of these bonds increases non-ideal solution behavior, which enhances the performance of the FPD.

ALTERNATIVE AIRCRAFT ANTI-ICING FORMULATIONS 2-2 • FPDs contain several oxygen atoms. This increases the hydrogen bonding and polar interactions with water, in addition to lowering the chemical’s BOD. • FPDs used on aircraft are typically nonionic compounds. Although various salts have been proposed, the intimate contact of FPD with the variety of materials and electrical systems found on aircraft tends to favor the use of non-ionic compounds. • Except for a carbonyl group, all the FPDs contain only single-bonded atoms. This is probably because of a requirement for high thermal stability and low chemical reactivity. • The use of amine groups in FPDs is limited. This is probably the result of the increased toxicity of amine-containing compounds and their potential for contributing to eutrophication of receiving waters. Using these general characteristics, a set of molecular structure constraints was developed to guide the search for candidate FPDs. The search was limited to those candidates whose molecular structure can be constructed from the following groups: >C< –H –O– –CO– Groups containing nitrogen were excluded because of concerns that they could contribute to eutrophication. Table 2-1 shows how the molecular structure of some of the FPDs listed above can be represented by these groups. TABLE 2-1. Structural group representation of FPDs. FPD Formula Groups 1,2-Propylene glycol C3H8O2 3 >C<, 8 –H, 2 –O– Ethyl lactate C5H10O3 4 >C<, 1 –CO–, 10 –H, 2 –O– Ethylene glycol C2H6O2 2 >C<, 6 –H, 2 –O– Glycerol C3H8O3 3 >C<, 8 –H, 3 –O– Triethylene glycol C6H14O4 6 >C<, 14 –H, 4 –O– Group representation enabled the computational generation of chemical structures for all candidate FPDs. The first step was to limit the number of carbons in a structure to a minimum of one and a maximum of six. A constraint also was imposed that at least two oxygen atoms be present in the structure. All possible combinations of groups satisfying these structural constraints were generated, and then the chemical formula for each of these combinations was generated. This procedure resulted in a total of 160 candidate chemical formulas. The following are examples of these candidate formulas: C2H2O2 C3H2O3 C3H6O3 C4H2O6 C4H10O4 C5H2O10 C5H12O5 C6H4O6 C6H14O6

SECTION 2—1BSUMMARY OF CANDIDATE DEICER COMPONENTS EVALUATED IN TEST PROGRAM 2-3 Each of these 160 candidate chemical formulas was used to search for commercially available chemicals using the National Institute of Standards and Technology Chemistry WebBook (5). The WebBook contains more than 70,000 chemical species. Candidates that contained chemical reactive groups such as acids or peroxides were excluded from consideration. Appendix A lists the 419 chemicals found in this search. Although any database search is limited to the extent of the database, the effectiveness of this search was validated because it found all the FPDs currently in use or being considered for use. Availability and Pricing Not all of the chemicals listed in Appendix A are available in commercial quantities. To determine commercial availability, the research team searched the online version of the Aldrich Chemical Catalog (6). The online catalog contains more than 40,000 commercially available chemicals. If a candidate was found in the online catalog, it was retained for further examination. If a candidate was not found in the online catalog, it was considered to be commercially unavailable and was rejected. This screening process yielded 219 candidate FPDs. Although some candidates were commercially available, their high price would prohibit use in a deicing or anti-icing fluid. Because these chemicals are available solely for research purposes, the prices in Aldrich Chemical Catalog are primarily for laboratory quantities and purities. These prices are often ten times greater than bulk prices. Because of these considerations, price was used very conservatively as a constraint. Only those chemicals whose price was greater than $750 per kilogram were eliminated from consideration. This screening step eliminated 87 candidates, leaving 132 for further consideration. Flash Point Paragraph 3.2.1 of the Aerospace Material Specification (AMS) 1424G states that fluids that are used as supplied must have a flash point not lower than 100°C. Although many deicing and anti-icing fluids are supplied mixed with water, this dilution has little effect on the FPD’s flash point. Figure 2-1 shows the flash points for methanol-water and ethanol-water mixtures (7). As Figure 2-1 shows, even at 60 percent water by weight, the flash points of the alcohol mixtures have increased by less than 20°C over the flash point of the pure alcohols. Using this observation in a conservative manner, the research team rejected all candidate FPDs with flash points below 75°C. This eliminated 79 candidates from further consideration. The flash points for 21 candidates were not available; however, these candidates all had high melting points, making them room-temperature solids. As such, their flash points would satisfy the constraint and they were included in the list of candidates for further consideration.

ALTERNATIVE AIRCRAFT ANTI-ICING FORMULATIONS 2-4 Figure 2-1. Flash points for mixtures of alcohol and water. 0 10 20 30 40 50 60 70 80 0 20 40 60 80 100 wt % Water Te m p [C ] Methanol Ethanol Biological Oxygen Demand BOD is a major environmental concern with deicers, and candidate FPDs were constrained to a BOD of the same general magnitude or lower than currently used FPDs. Quantifying this constraint is complicated by the effect of time, temperature, acclimation of microorganisms and extent of biodegradation on BOD test results. To overcome this difficulty, theoretical oxygen demand (ThOD) was used as the metric for quick screening. ThOD is the amount of oxygen needed to convert a chemical entirely into carbon dioxide, water, and other “final” combustion products. For candidate FPDs, ThOD can be computed by Equation 2-1. Mw OHCThOD )16832( −+= (2-1) In Equation 2-1, C is the number of carbon atoms in the candidate, H is the number of hydrogen atoms, and O is the number of oxygen atoms. The right-hand-side of the equation is equal to the grams of oxygen needed for complete oxidation per unit weight. The coefficients in Equation 2-1 give ThOD in units of grams of oxygen per gram of candidate. Table 2-2 gives the ThOD for several currently used FPDs. Considering the ThODs of current FPDs, all candidates with ThOD values greater than 2.0 were rejected. This criterion further reduced the list of candidates to 41. Freezing Point Depressant Paragraph 3.5.1 of AMS specification 1424G states that candidate deicing fluids must have a freezing point below -20°C when diluted 1:1 by volume with water. Figure 2-2 shows the freezing point curves for mixtures of ethylene glycol and water (8). The figure shows that mixtures with ethylene glycol concentrations between 36.5 and 94.8 percent will have freezing points below -20°C.

SECTION 2—1BSUMMARY OF CANDIDATE DEICER COMPONENTS EVALUATED IN TEST PROGRAM 2-5 TABLE 2-2. ThOD of some current FPDs. FPD ThOD (kg/kg) Glycerol 1.216 Ethylene glycol 1.288 DEG 1.508 Triethylene glycol 1.598 Propylene glycol 1.682 kg/kg = kilogram per kilogram Figure 2-2. Freezing point curves for mixtures of ethylene glycol and water. -40 -35 -30 -25 -20 -15 -10 -5 0 0 20 40 60 80 100 Weight % Ethylene Glycol Te m p [C ] Liquid Solid + Liquid Solid + Liquid The freezing point curves of candidate mixtures can be predicted from thermodynamics. If it is assumed that all candidate mixtures form simple eutectics and solid-solid phase transitions are ignored, then their freezing point curves are given by Equation 2-2. ( )       − ∆ −= m m TTR H x 11ln γ (2-2) In Equation 2-2, x is the mole fraction of the concentrated species, γ is the activity coefficient—a measure of solution non-ideality, ∆Hm is the enthalpy of fusion, R is the ideal gas constant, T is the initial melting point of the mixture, and Tm is the melting point of the pure compound. Equation 2-2 is only applicable for the concentrated portion of the phase diagram. To generate the complete phase diagram, Equation 2-2 is plotted for the concentrated water region first and then for the concentrated FPD region. Figure 2-3 shows such a construction. The intersection of the two curves denotes the eutectic point. The region above the two intersecting curves denotes the liquid phase. The region below the eutectic point, indicated

ALTERNATIVE AIRCRAFT ANTI-ICING FORMULATIONS 2-6 by the horizontal dashed line, denotes the solid phase. The regions below the curves but above the eutectic point indicate two phase regions—a mixture of solid and liquid phases. Figure 2-3. Construction of a hypothetical freezing point curve. -70 -60 -50 -40 -30 -20 -10 0 0 20 40 60 80 100 Weight % Te m p [C ] Freezing Point Curve Eutectic Point Freezing Point Curve This analysis was used as the basis for computationally evaluating the eutectic point for each of the candidate FPDs. Unfortunately, enthalpy of fusion values were found for only 16 candidates. Values were estimated for the remaining candidates, but review of these predictions raised concerns about their accuracy. As an alternative, a single, conservative value was chosen for the enthalpy of fusion in the analysis. This value was used to evaluate all candidates, including those for which experimental values were known. Figure 2-4 shows how varying the enthalpy of fusion affects a mixture’s eutectic point and thus its acceptability as a FPD. It was assumed that the FPD had a freezing point of 10°C and the solution exhibited ideal behavior. (For ease of computation, Figure 2-4 displays FPD concentration in mole percent.)

SECTION 2—1BSUMMARY OF CANDIDATE DEICER COMPONENTS EVALUATED IN TEST PROGRAM 2-7 Figure 2-4. Variation of eutectic point with variation in enthalpy of fusion (∆Hm is given in units of kJ/mol). -70 -60 -50 -40 -30 -20 -10 0 10 20 0.0 0.2 0.4 0.6 0.8 1.0 Mole % FPD Te m p [C ] Hm = 7.5 Hm = 12.5 Hm = 17.5 Ideal Freezing Point Curve Figure 2-4 shows that as the enthalpy of fusion increases, the mixture’s freezing point decreases. A lower value of the enthalpy of fusion is thus preferable in candidate FPDs. The lowest enthalpy of fusion values found for candidate FPDs was 7.5 kJ/mol. This was therefore chosen as the single, conservative value to use in the continued analysis. Using a value of 7.5 kJ/mol for the enthalpy of fusion, Figure 2-5 shows how the eutectic point varies as a function of FPD melting point. Figure 2-5. Variation of eutectic point with variation in FPD freezing point (A ∆Hm of 7.5 kJ/mol assumed). -70 -60 -50 -40 -30 -20 -10 0 10 20 0.0 0.2 0.4 0.6 0.8 1.0 Mole % FPD Te m p [C ] Tm = 0 C Tm = 50 C Tm = 100 C Ideal Freezing Point Curve Figure 2-5 shows that an FPD with a melting point of 100°C has a eutectic temperature near -30°C. This value satisfies the specifications. However, the liquid range for such an FPD is very narrow. This means that any preferential evaporation of water or FPD could move the

ALTERNATIVE AIRCRAFT ANTI-ICING FORMULATIONS 2-8 fluid into the two-phase, solid-liquid region, which could result in formation of residuals on the aircraft surface. From this analysis, an upper limit of 50°C was set on the freezing point of candidate FPDs. This freezing point results in a eutectic temperature near -40°C and a narrow, but acceptable, liquid range. Candidates with freezing points above this limit were eliminated from further consideration. Seven candidates did not have known freezing points. Values were estimated for these candidates using Joback’s group contribution method (9). Using these estimates and literature values for candidate melting points, the number of candidates was reduced to 27. Aquatic Toxicity One of the major goals of this project is to develop deicing and anti-icing fluids with lower aquatic toxicity than currently used products. Toxicity analyses conducted during this project identified the aquatic toxicity range for several current products (1). Tables 2-3 and 2-4 show the aquatic toxicity toward Ceriodaphnia dubia and Pimephales promelas, respectively. Products are designated by letter (A–N) to maintain anonymity. TABLE 2-3. 96-hour C. dubia LC50 of some current deicing/anti-icing products. Product Type Toxicity Limit [mg/l] 1 D Type I 33,977 2 C Type I 26,517 3 E Type I 11,468 4 A Type I 7,747 5 N Runway 4,547 6 K Type IV 2,600 7 N Runway 2,437 8 L Runway 1,302 9 H Type IV 948 10 I Type IV 662 11 J Type IV 528 LC50 = 50 percent lethal concentration. mg/L = milligrams per liter

SECTION 2—1BSUMMARY OF CANDIDATE DEICER COMPONENTS EVALUATED IN TEST PROGRAM 2-9 TABLE 2-4. 96-hour P. promelas LC50 of some current deicing/anti-icing products. Product Type Toxicity Limit [mg/l] 1 D Type I 32,256 2 C Type I 20,800 3 E Type I 12,172 4 A Type I 10,893 5 M Runway 6,757 6 N Runway 4,708 7 H Type IV 2,666 8 J Type IV 1,041 9 L Runway 960 10 K Type IV 888 11 I Type IV 219 Tables 2-3 and 2-4 are sorted in increasing order of toxicity. Sorting in this manner for the products investigated shows that Type I fluids have the lowest toxicity; Type IV fluids have the highest toxicity; and runway deicers are in between. This is a very interesting result because Type I and Type IV fluids often use the same FPDs, surfactants, and even corrosion inhibitors. To screen candidate components, the following toxicity goals were established for new complete formulations, based on the objective to improve toxicity over current products: • New Type I fluids will have a 96-hour LC50 in excess of 35,000 mg/L. • New runway deicing chemicals will have a 96-hour LC50 in excess of 7,500 mg/L. • New Type IV fluids will have a 96-hour LC50 in excess of 5,000 mg/L. These limits would be applicable to both Ceriodaphnia dubia and Pimephales promelas species. (Although these limits were established for screening purposes, the goal is to develop new products with the lowest toxicities possible.) If it is assumed that no synergistic or antagonistic effects occur between the components of a formulation, these formulation toxicity limits can be used to establish toxicity limits for pure, candidate FPDs. For example, a mixture containing a minimum of 38.9 wt% 1,2- propylene glycol in water is needed to meet the -20°C freezing point requirement. At this concentration, the toxicity limit of the FPD must be greater than 35,000 mg/L × 0.389 or 13,615 mg/L to satisfy SAE’s AMS 1424 Type I fluid toxicity constraint of 4,000 mg/L. 1,2-Propylene glycol has a reported 96-hour LC50 of 55,700 mg/l for Pimephales promelas (10).

ALTERNATIVE AIRCRAFT ANTI-ICING FORMULATIONS 2-10 Because this LC50 is greater than the toxicity constraint, 1,2-propylene glycol was retained for further consideration. Unfortunately, aquatic toxicity data were found for only 12 of the remaining 27 candidates. Two of these candidates had 48-hour LC50 values for carp below 280 mg/L (10). One of these candidates had a 96-hour LC50 value for bluegill of 90 mg/L (10). Although these species are different than those set in the constraints, they are representative of generally high toxicity and as a result were eliminated from further consideration. This lack of aquatic toxicity data clearly demonstrates the need for the experimental portion of the project. Mammalian Health Effects Deicing and anti-icing fluids may be accidentally inhaled or ingested and may come in contact with skin and eyes during application. As a result, the acute toxicity and irritation potential of the candidate FPDs must be low. For each of the remaining 24 candidates, health effect data from Sax’s handbook (11), the National Library of Medicine’s Hazardous Substances Data Bank (HSDB) website (12) and individual material safety data sheets (MSDSs) were examined. Analysis of these data eliminated 7 candidates based on oral toxicity values for rat and mouse species. Summary of Candidates Table 2-5 lists the 17 candidate FPDs that satisfy all of the selection constraints. Several of these candidates are commonly used in current deicing fluids. For example, 1,2-propylene glycol evaluated well against all selection constraints and is a major component in commercially available fluids. TABLE 2-5. Candidate FPDs recommended for further evaluation. Project ID Formula Candidate CAS RN 027.01 C3H4O3 Ethylene carbonate 96-49-1 039.03 C3H8O2 1,2-Propylene glycol 57-55-6 039.04 C3H8O2 1,3-Propylene glycol 504-63-2 040.01 C3H8O3 Glycerol 56-81-5 059.02 C4H6O3 Propylene carbonate 108-32-7 071.02 C4H10O2 2,3-Butanediol 513-85-9 071.04 C4H10O4 1,3-Butanediol 107-88-0 071.12 C4H10O2 2-Methyl-1,3-propanediol 2163-42-0 072.01 C4H10O3 Diethylene glycol 111-46-6 097.05 C5H8O2 4-Methyl-γ-butyrolactone 108-29-2 099.01 C5H8O4 Dimethyl malonate 108-59-8 111.02 C5H12O3 2-(2-Methoxyethoxy)-ethanol 111-77-3 145.01 C6H10O4 Dimethyl succinate 106-65-0 151.16 C6H12O3 2,2-Dimethyl-1,3-dioxolane-4-methanol 100-79-8

SECTION 2—1BSUMMARY OF CANDIDATE DEICER COMPONENTS EVALUATED IN TEST PROGRAM 2-11 TABLE 2-5. Candidate FPDs recommended for further evaluation. Project ID Formula Candidate CAS RN 157.01 C6H14O3 2-(2-Ethoxyethoxy)-ethanol 111-90-0 157.09 C6H14O3 Dipropylene glycol 25265-71-8 158.01 C6H14O4 Triethylene glycol 112-27-6 CAS RN = Chemical Abstracts Service Registry Number Appendix B provides detailed information for each of these candidate FPDs. Freezing Point Depressant Candidates for PDMs The following chemicals were identified in the literature as either current, past, or proposed FPDs for PDMs: 1,2-Propylene glycol 1,3-Propylene glycol Potassium formate Potassium acetate Sodium acetate Sodium formate Magnesium acetate Urea Calcium acetate Ethylene glycol Diethylene glycol Triethylene glycol Many of the same chemicals used to remove ice and snow from aircraft are also used to remove ice from runways. This is reasonable because similar constraints apply to both operations. The list above includes several metallic salts that are used for runway deicing. Using salts as FPDs has the advantage that each ion contributes on a molar basis to the lowering of the mixture’s freezing point. As a result, adding one mole of a monovalent salt to water actually contributes two moles of ions to the solution, assuming the salt completely dissociates. A search of the literature did not find runway deicing chemicals based on cations other than sodium, potassium, calcium, and magnesium anions. The use of other period 4 elements— specifically scandium, titanium, vanadium, chromium, manganese, iron, cobalt, nickel, copper, and zinc—was investigated. A search of the Alfa-Aesar Chemical Catalog (6) and Sax’s Dangerous Properties of Industrial Materials (11) found only organic salts of these elements with adverse health effects, high cost, or instability. Based on these observations, it was decided to limit the cations to sodium, potassium, calcium and magnesium. The anions found in PDMs include formates, acetates, lactates, and carbonates. All of these anions have a negatively charged oxygen molecule, which forms an ionic bond with the cation. This observation was used to generalize the search for PDMs whose molecular structure can be constructed from the following groups: >C< –H –O– –CO– KO– NaO– –OCaO– –OMgO– Groups containing nitrogen or phosphorus were not included because of environmental concerns. Groups containing silicon were not included because of health concerns. Groups

ALTERNATIVE AIRCRAFT ANTI-ICING FORMULATIONS 2-12 containing sulfur were not included because of corrosion concerns. Table 2-6 shows how the molecular structure of some of the FPDs listed above can be represented by these groups. TABLE 2-6. Structural group representation of FPDs potentially useful for PDMs. FPD Formula Groups 1,2-Propylene glycol C3H8O2 3 >C<, 8 –H, 2 –O– Potassium acetate C2H3KO2 1 >C<, 1 –CO–, 3 –H, 1 KO– Calcium pyruvate C6H6CaO6 2 >C<, 4 –CO–, 6 –H, 1 –OCaO– Magnesium lactate Mg(C3H5O3)2 4 >C<, 2 –CO–, 10 –H, 2 –O–, 1 –OMgO– Triethylene glycol C6H14O4 6 >C<, 14 –H, 4 –O– As before, the number of carbons in a structure was limited to a minimum of one and a maximum of six. At least two oxygen atoms were also required to be present in a structure. The number of metallic cations was limited to a minimum of one and a maximum of three per molecule. (One cation was required because non-ionic candidates have already been identified in the search for ADF and AAF FPDs.) All possible combinations of groups satisfying these structural constraints were generated, and then the chemical formula for each combination was generated. This procedure resulted in a total of 2,172 candidate chemical formulas. The following examples illustrate some of these candidate formulas: CHKO2 C2H2CaO4 C5H7NaO3 C4H5KO4 C6H6CaO6 C6H9NaO3 C5H4Na2O5 C4H6MgO4 C3Na2O5 Each of these 2,178 candidate chemical formulas was used to search for commercially available chemicals using the Aldrich Chemical Catalog (6) and the Alfa-Aesar Chemical Catalog (13). Candidates that contained chemical reactive groups such as acids or peroxides were excluded from consideration because of safety or material compatibility concerns. Appendix C lists the 59 chemicals found in this search. Although any database search is limited to the extent of the database, the search found all the FPDs currently in use or being considered for use in PDMs. Availability and Pricing Not all of the chemicals listed in Appendix C are available in commercial quantities at reasonable prices. To determine commercial availability the online version of the Aldrich Chemical Catalog (6) was searched. Again using price as a very conservative constraint, only those chemicals whose price was greater than $750 per kilogram were eliminated from consideration. This screening step eliminated 28 candidates and left 31 for further consideration. Another 132 candidates identified in the search for ADFs and AAFs also had acceptable pricing and were considered possible PDM candidates. These previously identified non- ionic candidates were combined with the current 31 ionic candidates to produce a compilation of 163 candidates that were investigated further.

SECTION 2—1BSUMMARY OF CANDIDATE DEICER COMPONENTS EVALUATED IN TEST PROGRAM 2-13 Flash Point Paragraph 3.2.1 of AMS specification 1435A states that runway deicing fluids must have a flash point not lower than 100°C. Paragraph 3.2.2 of AMS specification 1431B states that runway deicing solids must have a flash point not lower than 93°C. Because some of the candidates may be applied as mixtures with water, these constraints were relaxed slightly, and all candidate FPDs with flash points below 75°C were rejected. This eliminated 80 candidates from further consideration. The flash points for 37 candidates were not available. However, these candidates all had high melting points, making them room-temperature solids. It is therefore reasonable to assume that their flash points would satisfy the constraint, so they were included in the list of candidates for further consideration. Biological Oxygen Demand ThOD was again used to screen candidates for low BOD. Equation 2-1 shows that only carbon, hydrogen and oxygen were used in the calculation of ThOD. Metallic cations were assumed to make no contribution to ThOD. Twelve nonionic candidates with ThOD values greater than 2.0 were rejected. This reduced the current list of 83 candidates to 71. None of the ionic candidates were eliminated. Note on the Definition of BOD It is generally assumed that elements such as sodium, potassium, calcium, or magnesium are not oxidized by biochemical means. As a result, the amount of oxygen needed to degrade one molecule of sodium acetate is the same as the amount needed to degrade one molecule of potassium acetate. However, because potassium acetate has a higher molecular weight than sodium acetate, its ThOD is lower by 16 percent. Table 2-7 shows the calculations for these two chemicals. TABLE 2-7. ThOD values for FPDs. FPD Molecular Weight ThOD (g O2/g FPD) Sodium acetate 82.03 0.68 Potassium acetate 98.14 0.57 g O2/g FPD = gram of oxygen per gram of FPD By the same calculation, cesium acetate has a ThOD of 0.29 g O2 / g FPD even though on a molecular basis the same amount of oxygen is needed for degradation. A thorough comparison of runway deicing chemicals must also consider their ability to melt ice, penetrate ice, and disrupt the adhesion of ice to pavement. A deicing chemical’s freezing point depression is a major factor in all of these attributes. Therefore, comparing the freezing point depression of runway deicing chemicals should provide a fair prediction of actual performance. Table 2-8 shows the freezing points for aqueous solutions of sodium acetate and potassium acetate (14). Figure 2-6 shows the same data plotted on a molar basis.

ALTERNATIVE AIRCRAFT ANTI-ICING FORMULATIONS 2-14 TABLE 2-8. Freezing points of aqueous deicing chemical solutions. Weight % Sodium Acetate Potassium Acetate Mole % Tm [°C] Mole % Tm [°C] 5 1.1 -2.9 1.0 -0.2 10 2.4 -5.9 2.0 -5.0 15 3.7 -10.3 3.1 -7.8 20 5.2 -15.4 4.4 -11.5 25 — — 5.8 -15.9 30 — — 7.3 -22.7 Figure 2-6 shows that slightly more potassium acetate is needed than sodium acetate to achieve the same freezing point depression. For example, to melt 1 kilogram of ice at -15°C would require 2.98 moles of sodium acetate and 3.23 moles of potassium acetate. To oxidize these quantities of deicing fluid would take 166.9 grams of oxygen for sodium acetate and 180.9 grams of oxygen for potassium acetate. The sodium acetate requires less oxygen. This is the opposite conclusion that is drawn from comparing ThOD values. Figure 2-6. Freezing point depression on a molar basis. -25 -20 -15 -10 -5 0 0 2 4 6 8 10 Mol % Fr ee zi ng P oi nt [C ] Na Acetate K Acetate This analysis shows that simply comparing oxygen demand parameters is insufficient for evaluating the environmental impact of PDMs. Additional parameters such as molecular weight, freezing point depression, application rates, and runoff must be considered. The following metric is proposed for evaluating environmental impact: Grams of oxygen needed to oxidize the chemicals needed to achieve and maintain one square meter of runway free of ice for one hour.

SECTION 2—1BSUMMARY OF CANDIDATE DEICER COMPONENTS EVALUATED IN TEST PROGRAM 2-15 Melting, Penetration, and Undercutting Unlike aircraft deicing chemicals, runway deicers do not need to completely melt accumulated ice and snow. National Research Council publication SHRP-H-332 (15) details test procedures for three key properties of runway deicers: 1) the ability to melt ice, 2) the ability to penetrate ice, and 3) the ability to undercut ice. It was not possible to directly relate these properties to specific physical properties, but it is understood that a chemical’s freezing point depression is a key factor affecting each property. Freezing Point Depression AMS 1431B, the SAE specification for solid runway deicing chemicals, does not specify a required freezing point depression. However, manufacturers do specify a lowest use temperature. Sodium acetate can be used below -18°C. Potassium acetate can be used below -28°C. Sodium chloride should not be used below -10°C. Although sodium chloride is not used on airfields, it is used here as an example. Figure 2-7 shows the phase diagram for dilute solutions of sodium chloride in water. The figure shows that the recommended minimum use temperature of -10°C is approximately 10°C higher than the eutectic temperature. Figure 2-7. Phase diagram of aqueous sodium chloride. -30 -20 -10 0 10 20 30 0 2 4 6 8 10 12 Mole % NaCl Te m pe ra tu re [C ] Liquid and Solid Liquid and Solid Liquid Solid This observation was used to set a constraint that candidate FPDs for PDMs must have a eutectic temperature at least 10°C below their lowest use temperature. It was further decided to set the lowest use temperature for candidate runway deicers to a conservative value of -18°C. This results in the constraint that all candidate FPDs must have eutectic temperatures below -28°C. Because high melting point solids are acceptable runway deicing chemicals, the thermodynamic analysis used previously to screen for eutectic temperatures must be

ALTERNATIVE AIRCRAFT ANTI-ICING FORMULATIONS 2-16 modified. The approximate freezing point depression caused by adding any chemical to water is given by Equation 2-3. imT 86.1=∆ (2-3) In Equation 2-3, ∆T is the freezing point depression, i is the number of species and m is the molality of the solution. For a non-ionic compound, i is equal to 1. For an ionic compound, i is equal to the number of ionic species, assuming the compound is completely dissociated. For example, i for calcium chloride is three because it dissociates into one calcium ion and two chloride ions. Inserting the constraint on the eutectic temperature into Equation 2-4, the following constraint on FPD concentration is obtained. i m 05.15= (2-4) Table 2-9 shows the estimated concentrations needed to meet the use temperature constraint for several candidate FPDs. TABLE 2-9. Estimated concentrations for limiting eutectic temperature. Candidate FPD i [ions/mole] Concentration [mol/kg] Concentration [g/kg] Sodium formate 2 7.52 511.5 Potassium formate 2 7.52 632.4 Calcium formate 3 5.02 653.2 Tripotassium citrate 4 3.76 1164.3 Glucose 1 15.05 2711.4 Table 2-9 shows that a minimum of 2,711.4 grams of glucose must be mixed with 1,000 grams of water to produce a solution with an acceptable freezing point. Unfortunately, the solubility of glucose at 20°C is only 910 g/kg of water. Aqueous solutions of glucose cannot be formed that satisfy the limiting eutectic temperature constraint, so glucose would be eliminated from further consideration. Using this approach, the minimum concentration needed of each candidate to meet the eutectic temperature constraint was calculated. Solubility data were then obtained from Lange’s Handbook (16) and the HSDB Database (12). If the solubility limit was below the minimum concentration needed, the candidate was rejected. Twenty-four candidates were found to have unacceptable solubility and were rejected. Solubility data for 33 candidates could not be found. These were retained for further screening. Aquatic Toxicity The criterion for new runway-deicing formulations was established as a 96-hour LC50 in excess of 7,500 mg/L to provide a significant benefit over existing products. This limit would be applicable to both Ceriodaphnia dubia and Pimephales promelas species. (Although

SECTION 2—1BSUMMARY OF CANDIDATE DEICER COMPONENTS EVALUATED IN TEST PROGRAM 2-17 this limit was established for screening purposes, the goal is to develop new products with toxicities as low as possible.) Aquatic toxicity data were found for 19 of the remaining 47 candidates. Seven of these candidates had LC50 values below 7,500 mg/L and were eliminated from further consideration. Mammalian Health Effects Health effect data from Sax’s Handbook (11), the National Library of Medicine’s HSDB website (12) and individual MSDSs were examined for each of the remaining 40 candidates. Analysis of these data eliminated seven candidates because of high oral toxicity toward rats and mice. Carbon-Carbon Brake Oxidation and Runway Corrosion Many aircraft use carbon composite brake pads in their braking systems. During braking these pads become heated and reach temperatures in excess of 700°C. At these temperatures, carbon exposed to the air will quickly oxidize. To reduce such oxidation, non- wearing brake surfaces are coated with an antioxidant coating. Stover (17) describes an antioxidant coating composed of a zinc-aluminum-phosphate glass. This composition is similar to the generic aluminum-phosphate glass coating proposed by the G-12F Catalytic Oxidation Working Group. Walker et al. (18) describe the use of a phosphorus-containing undercoat covered with a boron-containing overcoat. Walker et al. (19) further describe a system in which the boron-containing overcoat also contains alkali or alkaline metal silicates, alkali metal hydroxides, boron nitride and boron carbide. Recent studies (20-22) of runway deicing chemicals indicate that sodium- and potassium- based fluids significantly degrade these antioxidant coatings, making carbon composite brakes susceptible to rapid high temperature oxidation. The chemical mechanism by which this degradation takes place is uncertain. Calcium- and magnesium-based fluids cause less degradation than sodium and potassium compounds. Non-ionic compounds do not harm antioxidant coatings. The current list of candidates contains 12 sodium-based chemicals and 9 potassium-based chemicals. Because considerable research is currently underway into the mechanism of antioxidant coating degradation and into new antioxidant coatings, it was decided to retain these candidates for further evaluation. Price Although the price of a chemical is highly dependent upon quantity, grade, and market conditions, the price constraint was tightened to reduce the number of candidate FPDs. Candidates with a price greater than $140 per kilogram were rejected. This constraint eliminated 11 candidates. Summary of Candidates Samples of potassium L-lactate could not be purchased. Because of this lack of availability, it was eliminated from further consideration. Table 2-10 lists the 21 candidate FPDs that

ALTERNATIVE AIRCRAFT ANTI-ICING FORMULATIONS 2-18 satisfy the selection constraints for new PDM formulations with improved environmental characteristics. Several of these candidates are used in current runway deicing products. TABLE 2-10. Candidate freezing point depressants recommended for further evaluation. Project ID Formula Candidate FPD CAS RN 27.01 C3H4O3 Ethylene carbonate 96-49-1 39.03 C3H8O2 1,2-Propylene glycol 57-55-6 39.04 C3H8O2 1,3-Propylene glycol 504-63-2 40.01 C3H8O3 Glycerol 56-81-5 71.04 C4H10O2 1,3-Butanediol 107-88-0 71.12 C4H10O2 2-Methyl-1,3-propanediol 2163-42-0 72.01 C4H10O3 Diethylene glycol 111-46-6 111.01 C5H12O3 1,1,1-Trimethanolethane 77-85-0 111.02 C5H12O3 2-(2-Methoxyethoxy)-ethanol 111-77-3 113.01 C5H12O5 Xylitol 87-99-0 147.03 C6H10O6 D-Gluconic acid, δ-lactone 90-80-2 151.16 C6H12O3 2,2-Dimethyl-1,3-dioxolane-4-methanol 100-79-8 157.01 C6H14O3 2-(2-Ethoxyethoxy)-ethanol 111-90-0 157.03 C6H14O3 Trimethylolpropane 77-99-6 157.09 C6H14O3 Dipropylene glycol 25265-71-8 158.01 C6H14O4 Triethylene glycol 112-27-6 332.01 C2H3NaO2 Sodium acetate 127-09-3 749.01 C6H10CaO4 Calcium propionate 4075-81-4 1509.01 C4H4Na2O4 Disodium succinate 150-90-3 1592.01 C4H4K2O6 L-Tartaric acid dipotassium salt 921-53-9 1829.01 C6H5K3O7 Tripotassium citrate 6100-05-6 The following 12 candidates were also recommended for use in aircraft deicing and anti- icing fluids: • 2,2-Dimethyl-1,3-dioxolane-4-methanol • Dipropylene glycol • 2-(2-Methoxyethoxy)-ethanol • Ethylene carbonate • 2-(2-Ethoxyethoxy)-ethanol • Triethylene glycol • 2-Methyl-1,3-propanediol • Diethylene glycol • 1,2-Propylene glycol • 1,3-Butanediol • 1,3-Propylene glycol • Glycerol Appendix D details the known information for each of the remaining 9 candidate FPDs.

SECTION 2—1BSUMMARY OF CANDIDATE DEICER COMPONENTS EVALUATED IN TEST PROGRAM 2-19 Thickener Candidates Figure 2-8 shows the rheological behavior (flow of fluids under strain) of a typical anti-icing fluid. At low shear rates, the fluid has a high viscosity, promoting the formation of a thick coating and slowing drainage. At high shear rates, such as those experienced during takeoff, the fluid has a low viscosity, causing it to readily shed from aerodynamic surfaces. Figure 2-8. Rheological behavior of typical anti-icing fluid at 0°C. 100 1000 10000 100000 0.01 0.1 1 10 100 Shear Rate [1/sec] Vi sc os ity [c P] The shear-thinning behavior required of anti-icing fluids is typically achieved by using associative or particulate thickeners. Typical associative thickeners include polysaccharides such as xanthan gum, welan gum, or carrageenan gum. The most common particulate thickener is lightly cross-linked polyacrylic acid. Associative thickeners are long-chain, water-soluble polymers. In solution, these polymers are fully extended and randomly oriented. At concentrations greater than what is termed the “overlap concentration,” parts of the polymer chain will associate or entangle. These entanglements produce the high viscosity observed in these polymer solutions. However, these associations also are very weak and can be easily broken by molecular motion. This ease of breakage is why associative thickeners exhibit dramatic shear-thinning behavior. Polyacrylic acid thickeners typically consist of lightly cross-linked particles having diameters less than 1 micron. When placed in an aqueous solution having a pH in the range of 4 to 10, these particles can swell to 500 times their dry volume. The high viscosity of these solutions is caused by the presence of these large particles and possibly the entanglement of polymeric chains emanating from the particle’s surface. Because these particles are lightly cross-linked, they are easily deformed by fluid motion. This ease of deformation is why particulate thickeners exhibit dramatic shear-thinning behavior. Thickener Candidates Four categories of thickening agents that can produce the shear-thinning rheological behavior needed in anti-icing fluids are:

ALTERNATIVE AIRCRAFT ANTI-ICING FORMULATIONS 2-20 1. Polysaccharides: These agents are long-chain polymers comprising repeat sugar units. They are widely used to thicken food products. Several polysaccharides produce highly shear-thinning aqueous solutions that rapidly recover viscosity after shearing. 2. Synthetic Polymers: Many long-chain polymers will increase the viscosity of aqueous solutions. If this viscosity results from weak inter-molecular bonds or steric hindrance, it is very likely the solution will exhibit shear-thinning behavior. 3. Organoclays: These agents are commonly based on hectorite, a hydrophilic swelling clay composed of silicate sheets that delaminate in water. Organoclays are widely used as rheological modifiers in waterborne coatings, sealants, inks, paper coatings, and ceramics. Organoclay solutions are thixotropic—when a shear is applied the solution’s viscosity is greatly reduced; when the shear is removed the solution’s viscosity slowly returns to its original high value. 4. Fumed Silica: This agent is composed of small silicon dioxide particles having surface hydroxyl groups. In solution these hydroxyl groups hydrogen bond with groups on other particles, creating a weak network. This network is easily disrupted by applied shear and will reform once the shear is removed. Organoclays and fumed silica were not considered viable candidates because concern about the effects of adding dispersed solids (potential for settling of solids or residuals) to anti- icing fluids outweighs any possible environmental and performance improvements. Polysaccharide and synthetic polymer candidates were investigated further. Polysaccharides Polysaccharides are polymers typically composed of 40 to 3,000 repeat sugar units. Polysaccharides are found in seaweeds, seeds, and plant exudates and are produced by many microorganisms. The most common industrial polysaccharides are (23): 1. Agar: a hydrophilic colloid extracted from marine algae of the class Rhodophyceae. It is insoluble in cold water but soluble in boiling water. Agar gels are used in microbiological research, to stabilize food products, and in many medicinal products. The viscosity of agar solutions is too low for use in anti-icing fluids (24). 2. Alginates: occurs as alginic acid in concentrations ranging from 18 to 40 percent in all brown seaweeds (23). The sodium, potassium, ammonium or propylene glycol salts of alginic acid are typically used as thickening agents. The ester linkages in propylene glycol alginate prohibits it use in alkaline solutions. Although alginate salts can produce aqueous solutions with very high viscosity, their shear-thinning behavior is not large enough for use in anti-icing fluids. 3. Carrageenan: extracted from a number of red seaweeds. It is commonly used as a thickener in toothpastes, shampoos and pharmaceuticals. The viscosity of carrageenan solutions is too low for use in anti-icing fluids (25). 4. Guar Gum: obtained from the seed of the legume Cyamopsis tetragonolobus, an annual plant that grows mainly in the arid and semiarid regions of India. Although guar gum can produce high viscosity, shear-thinning solutions, the amount of shear-thinning is not large enough for use in anti-icing fluids (23).

SECTION 2—1BSUMMARY OF CANDIDATE DEICER COMPONENTS EVALUATED IN TEST PROGRAM 2-21 5. Locust Bean Gum: obtained from the seed of the carob tree, which grows in several Mediterranean countries. Used by the ancient Egyptians to bind mummies, locust bean gum is a neutral polysaccharide with a molecular weight reported to be between 300,000 and 360,000 daltons. Locust bean gum solutions are reported to have lower viscosities than carrageenan solutions. Therefore they are not considered as viable candidate thickeners. 6. Pectin: found in all higher plants. Widely used to form food gels. Pectin’s solution viscosity is too low for use in anti-icing fluids. 7. Xanthan Gum: an exocellular heteropolysaccharide produced by the Xanthomonas campestris bacterium. It is an approved food additive, commonly used in salad dressings and ice creams. Xanthan gum is used in industrial products to thicken fluids, especially as a thickener for drilling fluids. Studies indicate the molecular weight of xanthan gum to be approximately two million daltons (26). Aqueous solutions of xanthan gum have high viscosity and are highly shear-thinning. A 0.5 wt% solution has a viscosity of 8,000 centipoise (cP) at a shear rate of 1 sec-1 and a viscosity of 200 cP at a shear rate of 100 sec-1 (23). The U.S. Environmental Protection Agency’s (EPA’s) ECOTOX database reports a 96-hour LC50 of 420 mg/L for xanthan gum toward rainbow trout. Xanthan gum is recommended for further investigation. 8. Welan Gum: produced by an Alcaligenes species of bacteria in aerobic fermentation. Welan gum solutions have exceptional stability at temperatures of 100°C for extended periods of time (23). Aqueous solutions of welan gum exhibit significant shear-thinning behavior. A 1.0 wt% solution has a viscosity of approximately 10,000 cP at a shear rate of 0.1 sec-1 and a viscosity of 1000 cP at a shear rate of 1.7 sec-1. Welan gum is recommended for further investigation. 9. Rhamsan Gum: an anionic, extracellular, microbial polysaccharide produced by a strain of Alcaligenes bacteria under aerobic fermentation conditions. Although rhamsan gum can produce high viscosity, shear-thinning solutions, the amount of shear-thinning is not large enough for use in anti-icing fluids (23). 10. Methylcellulose: a methyl ether of cellulose formed from substituting the hydrogen on some of cellulose’s hydroxyl groups with methyl groups. Methylcellulose exhibits inverse solubility—it is soluble in cold water but insoluble in warm water. Methylcellulose is available in a wide range of molecular weights, from 10,000 to 250,000 daltons. Although it can produce high viscosity, shear-thinning solutions ( a 2 wt% solution of an 86,000 molecular weight methylcellulose has a viscosity of 6,000 cP at a shear rate of 1 sec-1 and 1,300 cP at a shear rate of 100 sec-1), the amount of shear- thinning is not large enough for use in anti-icing fluids (23). 11. Hydroxyalkyl Cellulose: a family of non-ionic cellulose ether polymers that are readily soluble in water and are produced in a wide range of molecular weights. Hydroxyethylcellulose is the most widely used member of this family. It is a common thickener for latex paints. A 2 wt% solution has a nominal viscosity of 50,000 cP and is moderately shear-thinning (23). Although hydroxyethylcellulose’s properties are not outstanding, it is recommended for further investigation because it is the only non-ionic

ALTERNATIVE AIRCRAFT ANTI-ICING FORMULATIONS 2-22 thickener with acceptable properties. Experimental evaluation of hydroxyethylcellulose may help determine if ionic nature influences aquatic toxicity. Synthetic Polymers Many synthetic polymers are used to thicken cosmetics, paints, coatings, and personal care products. Some of the more common polymers are: 1. Cross-linked Polyacrylic Acid: typically used as a lightly cross-linked polymer in anti- icing fluids, detergents, and surface cleaners. Polyacrylic acid thickeners typically consist of lightly cross-linked particles having diameters less than one micron. When placed in an aqueous solution having a pH in the range of 4 to 10, these particles can swell to 500 times their dry volume. The high viscosity of these solutions is caused by the presence of these large particles and possibly the entanglement of polymeric chains emanating from the particle’s surface. Because these particles are lightly cross-linked they are easily deformed by fluid motion. This ease of deformation is why particulate thickeners exhibit dramatic shear-thinning behavior. 2. Polyethylene Oxide: a water soluble, non-toxic, non-ionic polymer used in paints, pharmaceuticals, cleaners, plasticizers, heat transfer fluids, and paper coatings. Also known as polyethylene glycol. Typical molecular weights range from 200 to 7000 daltons. A 50 percent aqueous solution of a 6,000-dalton polyethylene oxide has a viscosity of 100 cP. Polyethylene oxide’s solution viscosity is too low for use in anti-icing fluids. 3. Polyvinylpyrrolidone: a water-soluble homopolymer used in the formulation of cosmetics, adhesives and as a plasma extender. A 5 wt% solution of 3,000,000 molecular weight polyvinylpyrollidone was reported to have a viscosity of 350 cP at 25°C. Polyvinylpyrrolidone’s solution viscosity is too low for use in anti-icing fluids. Table 2-11 lists the five thickeners selected for experimental evaluation. TABLE 2-11. Candidate thickeners. Project ID Candidate Chemical Class 02.001 Kelzan-HP Polysaccharide 02.002 Kelzan-RD Polysaccharide 02.003 K1A96 Polysaccharide 02.004 Cellosize DCS HV Modified Polysaccharide 02.005 Carbopol EZ-4 Synthetic Polymer Surfactant Candidates If a mixture of water and propylene glycol was poured onto an aircraft surface, most of the liquid would either bead up into drops or roll right off the surface. This behavior is caused

SECTION 2—1BSUMMARY OF CANDIDATE DEICER COMPONENTS EVALUATED IN TEST PROGRAM 2-23 by the mixture’s high surface energy—creating a new surface requires considerable energy, so the liquid attempts to minimize its surface area by forming a spherical drop. Surfactants are added to aircraft deicing and anti-icing fluids to reduce their surface energy. With lower surface energy, these fluids will completely spread across aircraft surfaces, thereby ensuring an unbroken liquid coating. Environmental concerns regarding surfactants are primarily due to aquatic toxicity. The low percentage of surfactant in overall deicer formulations does not pose a significant BOD concern. Most of the surfactants that were examined have LC50 values ranging from 1 to 10 mg/L towards fish species. Some have values ranging from 10 to 200 mg/L. Many surfactants lack any experimental values for aquatic toxicity. Several general observations can be made regarding surfactant toxicity: • The toxicity of alcohol ethoxylates generally decreases with increasing ethylene oxide chain length. • Branched alkyl chains are less toxic than linear alkyl chains. • Secondary alcohols are less toxic than primary alcohols. • Surfactants containing mixtures of ethylene oxide and propylene oxide are less toxic than those containing only ethylene oxide. Unfortunately, using these observations to select surfactants with lower aquatic toxicity may decrease performance and biodegradability. Longer ethylene oxide chain lengths increase a surfactant’s water solubility, which often causes more foaming. Longer chain lengths also increase surface viscosity, which increases foam stability. Branched alkyl chains degrade slower than linear alkyl chains. Similarly, propylene oxide chains degrade slower than ethylene oxide chains. These potential reductions in performance and biodegradability must be compared against reductions in aquatic toxicity. Several performance properties were used to select surfactants. The hydrophilic-lipophilic balance (HLB) is a measure of a surfactant’s water affinity to its oil affinity. The critical micelle concentration (CMC) is the concentration at which a surfactant saturates water’s surface. The cloud point is a measure of surfactant solubility. Of these performance properties, the HLB was the predominant factor used in selection. Surfactant molecules contain two structural fragments that possess differing solubility. For example, the structure of nonyl ethoxylate is The alkyl fragment on the left side of the molecule has high oil solubility, whereas the ethylene oxide fragment on the right side of the molecule has high water solubility. This binary nature of surfactants can be quantified into a number called the HLB. HLB values depend upon the type and size of each side of the surfactant molecule. Values are typically scaled to fall within the range of 1 to 20. Surfactants with high HLB values are very hydrophilic and promote oil-in-water emulsions. High-HLB surfactants also have a O(CH 2 CH 2 O) y H CH 3 (CH 2 ) 7 CH 2

ALTERNATIVE AIRCRAFT ANTI-ICING FORMULATIONS 2-24 tendency to create stable foams. Surfactants with low HLB values are very lipophilic and promote water-in-oil emulsions. Surfactants with intermediate HLB values can promote either type of emulsion but generally promote oil-in-water emulsions. Surfactants were selected with HLB values in the range of 10 to 15. These should promote wetting of aircraft surfaces while minimizing foaming. Figure 2-9 shows a graph of surface tension versus surfactant concentration. At zero concentration, the surface tension is equal to that of pure water. As surfactant is added, the surface tension decreases. At a certain concentration, the CMC, adding more surfactant does not cause any further decrease in the surface tension. Figure 2-9. Determination of critical micelle concentration. 20 30 40 50 60 70 80 0 0.1 0.2 0.3 0.4 0.5 Concentration [wt%] Su rf ac e Te ns io n [d yn es /c m ] The CMC is a measure of the efficiency of a surfactant. If two surfactants both produce the same minimum surface tension value in water, the surfactant with the lower CMC will be more efficient because it produces the same low surface tension with a lower concentration. The cloud point is the temperature above which an aqueous solution of a water-soluble surfactant becomes turbid. Storing a surfactant mixture at temperatures significantly higher than its cloud point may result in phase separation. In general, non-ionic surfactants show optimal effectiveness when used near or below their cloud point. Low-foam surfactants should be used at temperatures slightly above their cloud point. Table 2-12 presents an extensive list of the chemical categories of non-ionic surfactants currently available. For each category, the research team investigated three properties: performance, toxicity, and biodegradability. Performance includes properties such as water solubility, surface tension reduction, HLB, and chemical stability. Table 2-12 indicates which surfactant categories fail to contain acceptable candidates (27, 28). Categories that are believed to contain promising candidates are investigated in more detail. TABLE 2-12. Non-ionic surfactant categories. Surfactant Category Performance Toxicity Biodegradability Acetylenic diols — — — Alkolamides/alkanolamides Faila — —

SECTION 2—1BSUMMARY OF CANDIDATE DEICER COMPONENTS EVALUATED IN TEST PROGRAM 2-25 TABLE 2-12. Non-ionic surfactant categories. Surfactant Category Performance Toxicity Biodegradability Alkoxylated alkyl phenols — Failb Failb Alkoxylated branched alcohols — — — Alkoxylated linear alcohols — — — Alkoxylated secondary alcohols — — — Alkyl dimethylamine oxides Faila — — Alkyl glucamides Failc — — Alkyl polyglucosides — — — Amine EO-PO copolymers — — — Dialkyl dimethyl polysiloxanes — — — EO-PO copolymers — — — Esterified EO-PO copolymers Failc — — Ether amine oxides Faila — — Ethoxylated alkanolamides — — — Ethoxylated alkyl phenols — Failb Failb Ethoxylated castor oil Failc — — Ethoxylated ether amines — — — Ethoxylated fatty acids Failc Faild Faild Ethoxylated fatty amines — — — Ethoxylated phenol — Failb Failb Ethoxylated sorbitan esters Failc — — Fatty acid esters Failc — — Fatty alcohol EO-PO copolymers Failc — — Fatty amine oxides Failc — — Fluorinated alkyl alkoxylates — — Faile Fluorinated alkyl esters — — Faile Fluorinated alkyl polyoxyethylene ethanols — — Faile Glycerol esters Failc — — Sorbitan esters Failc — — aProduce large amounts of stable foam. bPresence of the aromatic ring increases toxicity and reduced biodegradability. cThese surfactants hydrolyze in alkaline formulations. dThese surfactants readily hydrolyze under acidic or alkaline conditions. eThe fluorinated portion of the surfactant does not degrade. Table 2-12 shows that 10 surfactant categories contain potentially promising candidates. Eighteen surfactants from these categories were selected for experimental testing. Appendix

ALTERNATIVE AIRCRAFT ANTI-ICING FORMULATIONS 2-26 E lists detailed information on these surfactant candidates. Samples of Pluronic L44 NF could not be obtained and so it was dropped from further consideration. Certain surfactant-FPD mixtures can produce foam when applied to aircraft surfaces. This foam is very undesirable because it may hinder the “clean wing” decision needed for takeoff. Antifoams are chemicals that can be added to ADFs to facilitate the drainage and breakage of foams. Unfortunately, antifoams are known to have low aquatic toxicity limits and thus should be used sparingly. It was decided to test antifoams with only two candidate surfactants to assess the need for including them in the final formulated product. Tergitol TMN-10 is an alkoxylated branched alcohol that is reported to have good physical and toxicological properties but may produce too much foam for the test application. Triton CG-110 is an alkyl polyglucoside that is reported to produce a significant quantity of foam. If an antifoam could successfully control foaming with these two surfactants, it should be similarly effective with any of the other surfactants. Table 2-13 lists the 17 surfactants and two mixtures selected for experimental evaluation. TABLE 2-13. Candidate surfactants. Project ID Trade Name Description 03.001 Surfynol 465 Acetylenic diol 03.002 Tergitol TMN-6 Alkoxylated branched alcohol 03.003 Tergitol TMN-10 Alkoxylated branched alcohol 03.004 Lutensol XP 50 Alkoxylated branched alcohol 03.005 Lutensol XP 100 Alkoxylated branched alcohol 03.006 Triton DF-16 Alkoxylated linear alcohol 03.007 Bio-Soft N1-5 Alkoxylated linear alcohol 03.008 Bio-Soft N1-7 Alkoxylated linear alcohol 03.009 Merpol SE Alkoxylated linear alcohol 03.010 Lutensol TDA 10 Alkoxylated linear alcohol 03.011 Plurafac S-405LF Alkoxylated linear alcohol 03.012 Tergitol 15-S-7 Alkoxylated secondary alcohol 03.013 Tergitol 15-S-12 Alkoxylated secondary alcohol 03.014 Triton CG-110 Alkyl polyglucoside 03.015 Tetronic 904 Amine ethylene-oxide propylene-oxide copolymer 03.016 Tergitol L-64 Ethylene-oxide propylene-oxide copolymer 03.018 Triton CF-32 Ethoxylated alkanolamide 03.019 - mixture - Tergitol TMN-10 + 10% Ridafoam NS 221. 03.020 - mixture - Triton CG-110 + 10% Ridafoam NS 221.

SECTION 2—1BSUMMARY OF CANDIDATE DEICER COMPONENTS EVALUATED IN TEST PROGRAM 2-27 Corrosion Inhibitor Candidates The corrosion of a metal requires oxygen, a conductive solution, and surface areas with differing chemical potential. Because most deicing and anti-icing chemicals contain salts or are mixed with water that contains salts, corrosion of steel and aluminum can readily occur. All deicing and anti-icing chemicals contain one or more corrosion inhibitors. Most corrosion inhibitors function by forming a physical or chemical coating on a metallic surface to prevent contact with oxygen and conductive solutions. There are many types of corrosion inhibitors for various applications and they can be grouped into two broad categories: organic and inorganic. As a class, organic corrosion inhibitors have less environmental impact than inorganic inhibitors. This observation resulted in a general focus on the selection of organic corrosion inhibitors, although several solid, inorganic inhibitors are recommended for testing in runway deicing chemicals. Most organic corrosion inhibitors form physical or chemical coatings on metallic surfaces. This surface activity explains why many organic corrosion inhibitors also act as surfactants. The hydrophiles for organic corrosion inhibitors include amines, acids, esters and alcohols. These hydrophiles adsorb onto the metallic surface, leaving their hydrophobic substructures to form a repellant coating. Corrosion inhibitor packages representative of current deicer products are available from PMC Specialties Group, Inc. under the Cobratec brand. These include: • Cobratec TT-50-S: an alkaline solution of sodium tolyltriazole dissolved in water • Cobratec TT-100: solid tolyltriazole available in either prill or powder forms Candidate corrosion inhibitors were selected primarily based on prior experience. Documentation of corrosion inhibition toward steel and aluminum was the first criterion. Aquatic toxicity was the second criterion. Quantitative data for these criteria were often not available, so qualitative information had to be used for selection. Appendix F lists 17 candidate corrosion inhibitors that were recommended for further investigation. Anti-Caking Additive Candidates Sodium formate is an effective pavement deicing chemical possessing significantly lower BOD than alternative solid FPDs. However, a disadvantage of this compound is that because of its hygroscopic nature, its granules tend to cake, which makes handling and dispersion difficult. During a period of elevated humidity, sodium formate will form an adsorbed layer of water on particle surfaces. Some of the salt will dissolve into this layer, forming a concentrated solution. When the humidity is reduced or the temperature increased, the adsorbed water evaporates and the dissolved salt recrystallizes. These recrystallized salts often form bridges or welds between particles. This process is called caking and results in the inability of particles to flow freely.

ALTERNATIVE AIRCRAFT ANTI-ICING FORMULATIONS 2-28 Two additives that have been used or have been proposed for preventing the caking of formate particles are: • Potassium carbonate: (29) describe using less than 5 percent by weight of potassium carbonate to eliminate caking. They also discuss the use of sucrose and mannose as anti- caking additives. • Tripotassium citrate: (30) describes the use of 10 percent by weight tripotassium citrate to eliminate caking. Both of these additives were evaluated in Tier 2.

Next: Section 3: Experimental Plan »
Alternative Aircraft Anti-Icing Formulations with Reduced Aquatic Toxicity and Biochemical Oxygen Demand Get This Book
×
 Alternative Aircraft Anti-Icing Formulations with Reduced Aquatic Toxicity and Biochemical Oxygen Demand
MyNAP members save 10% online.
Login or Register to save!
Download Free PDF

TRB’s Airport Cooperative Research Program (ACRP) Web-Only Document 8: Alternative Aircraft Anti-Icing Formulations with Reduced Aquatic Toxicity and Biochemical Oxygen Demand explores the aquatic toxicity and biological oxygen demand state of the art, components, and promising alternative formulations of deicing and anti-icing products. The report also examines the performance; efficiency; material compatibility; and environmental, operational, and safety impacts of alternative formulations and components as well as the fate and transport of deicing and anti-icing formulation components and their degradation products.

A summary of this report was produced as ACRP Research Results Digest 9.

READ FREE ONLINE

  1. ×

    Welcome to OpenBook!

    You're looking at OpenBook, NAP.edu's online reading room since 1999. Based on feedback from you, our users, we've made some improvements that make it easier than ever to read thousands of publications on our website.

    Do you want to take a quick tour of the OpenBook's features?

    No Thanks Take a Tour »
  2. ×

    Show this book's table of contents, where you can jump to any chapter by name.

    « Back Next »
  3. ×

    ...or use these buttons to go back to the previous chapter or skip to the next one.

    « Back Next »
  4. ×

    Jump up to the previous page or down to the next one. Also, you can type in a page number and press Enter to go directly to that page in the book.

    « Back Next »
  5. ×

    To search the entire text of this book, type in your search term here and press Enter.

    « Back Next »
  6. ×

    Share a link to this book page on your preferred social network or via email.

    « Back Next »
  7. ×

    View our suggested citation for this chapter.

    « Back Next »
  8. ×

    Ready to take your reading offline? Click here to buy this book in print or download it as a free PDF, if available.

    « Back Next »
Stay Connected!