Identification and Characterization of Nonflammable Azeotropic Mixtures (Flash Point)

Flash Point A flash point is a minimum temperature at which a flammable substance’s vapor will ignite in the presence of an ignition source, and rapidly lead the flame back to the source of the vapor. The process of rapidly leading the flame back to the source of the vapor is referred to as “flashing”, hence the name “flash point”. At this temperature, the vapor pressure of the substance is high enough to create a ratio of flammable molecules to oxygen in the air that overcomes the LEL of the substance. To make the flash point measurement as objective as possible, the ASTM D56 method is very clear about what constitutes a flash, because the operator must see the flash event (which occurs very quickly) and judge if it was a flash or not. Also, some substances participate in burning by making the test flame larger, changing its color, or producing a halo around it, while not actually exhibiting a flash. Flash point is not the only tool for determining flammability. The fact that a substance does not flash, does not mean that the substance will not burn. However, flash point is a good method for determining if a substance is safe for use in a vapor degreaser. There are different methods of flash point testing for different types of materials. There is open-cup flash point, which is used for substances with low volatility, and closed-cup flash point, which is used for substances with high volatility. Within these categories, there are tests that deal with large volumes (such as liters or more), small volumes (such as microliters or less), different types of ignition sources, different methods of heating the sample, with stirring, without stirring and many more variations. In this study, a closed-cup flash point tester with an electric bath heater and a natural gas ignition flame was used. More information about the flash point tester, procedure, and reasoning for choices made for this test can be found in other Chapter II and Chapter IV of this thesis. Distillation Theory A simple distillation of a mixture of two liquids will yield a distillate that is enriched in the component of the mixture that has the lower boiling point. A fractional distillation is a little more complex. A fractional distillation will yield the pure component with the lower boiling point in most cases, because of the long fractionating column that is used. This fractionating column works as if many simple distillations are occurring as the distillate makes its way up the column. The hotter-to-cooler temperature gradient in the column allows the sample to vaporize and move up the column then condense on a cooler part of the column to run down into a hotter part which vaporizes the sample again. In most cases, more than enough of these simple distillation steps occur to purify the lowest boiling component as the distillate in a fractional distillation. The cases where the distillate is still composed of more than one substance after fractional distillation are deemed low-boiling azeotropic mixtures or near-azeotropic mixtures. When looking at a temperature versus composition phase diagram, which can be seen in Figure 5, a distillation will always yield a distillate that is down-hill on the curve from the starting point. Figure 5. Temperature versus composition phase diagram of a non-azeotropic mixture. In Figure 5, Tb1 and Tb2 represent the boiling points of components one and two respectively. One distillation step shows what happens to the composition if a simple distillation is performed. For vaporization of the mixture one follows the horizontal line, and for condensation the vertical line. In a temperature versus composition phase diagram of a low-boiling azeotrope, the azeotropic point is down-hill from everything. So, the azeotrope will be collected as the distillate. If a mixture is collected as the distillate, and it is not an azeotrope, the mixture would be near-azeotropic. Hydrostatic Density Density determination using a DuNouy Tensiometer is quite simplistic. The mass of a Pyrex standard density sinker is taken. This is referred to as the dry mass (md), and the density of the sinker is represented by ρ1. The sinker is submerged in the liquid whose density (ρ2) is to be determined, and the sinkers apparent mass is taken. This is referred to as the wet mass (mw). These measurements are discussed in more detail in Chapter II and the Appendix. The density of the air (ρ3) is also used, but is relatively constant in a temperature controlled room. The setup of this experiment is shown in Figure 6. Figure 6. Diagram of a hydrostatic density measurement. Once all of this information is gathered, the density of the liquid can be calculated using eq 10.15 (10) Because density changes with temperature, and surface tension and viscosity depend on the density, the density measurement will need to be interpolated or extrapolated to the temperatures of those measurements. The interpolation or extrapolation can be performed using the thermal expansion coefficient (β). This coefficient can be calculated for a small temperature range as seen in eq 11, and its derivation and the derivation of its relation to density is discussed elsewhere.13,16 (11) In equation 11, ρ0 and T0 are the density and temperature at the lower temperature, and ρ1 and T1 are the density and temperature at the higher temperature. Surface Tension The experimental steps taken during a surface tension measurement with a DuNouy ring tensiometer are simple, however, the calculations needed to get to the final surface tension are quite long compared to the density calculations. A diagram of the tensiometer can be seen in Figure 11 in Chapter II. In the measurement, a platinumiridium ring is submerged in the liquid sample, and the liquid sample is slowly lowered until the ring begins to rise out of the surface. When the ring rises out of the surface of the liquid, it lifts the meniscus. The basic concept of the measurement is that the maximum amount of liquid lifted by the ring is proportional to the surface tension of the liquid. This lifting of the meniscus can be seen in Figure 7. The figure was used with permission from reference 14. Figure 7. Tensiometer ring lifting the meniscus of the sample.15 This measurement produces a force curve such as the one shown below in Figure 8. Figure 8. Force curve example. The mass lifted reaches a maximum value, then the mass starts to decline as the meniscus thins before it breaks free from the ring. The maximum mass lifted (m) is used to calculate the maximum liquid volume lifted (V) in eq 12. Equations 12-20 were taken from reference 14.15 (12) The maximum mass lifted can also be used to calculate the maximum equilibrium force of detachment (F) as seen in eq 13. (13) In eq 13, g is the gravitational acceleration (g = 980.665 cm/s2). This force and the radius of the ring (R) can be used to calculate a first approximation of the surface tension (γ) as seen in eq 14. (14) Two consecutive corrections are applied to this approximation. The sources of these corrections are discussed further elsewhere.15 The first correction term is for the interface between phase 2 and phase 3, the liquid and the air. This correction (ƒ23) is a function of R3/V and R/a where a is the wire radius of the ring. This correction term is applied as seen in eq 15. ƒ (15) The second correction term is calculated using the shape of the interface between phases 2 and 3. The corrected surface tension from eq 15 is used to calculate a scaling factor (C23), seen in eq 16, for R and V. (16) This scaling factor is applied to the radius of the ring as seen in eq 17. (17) The scaling factor is applied to the volume lifted as seen in eq 18. / (18) R’ and V’ are used to calculate the second correction to the surface tension (ƒ’) as seen in eq 19. ƒ (19) The final surface tension value is calculated using eq 20. ƒ′ (20) It is recommended to the reader to do these calculations in a spreadsheet, especially if the calculations need to be done more than once. Viscosity Viscosity can be defined as a measure of a substance’s resistance to flow. This property can be measured in a few different ways. However, the ball-drop method was used in this study. The experimental steps for this method are discussed in Chapter II. The main concept of the experiment is that a ball of known density is allowed to fall through a tube whose dimensions are well defined and is filled with the sample whose density is also known. When the ball is falling to the bottom of the tube, it forces the liquid sample to flow around it. This flow rate is measured by the amount of time that the ball takes to fall through a predetermined timing zone. Equations 21 and 22 are taken from ASTM D1343.17 Equation 21 is normally used to calculate a term known as the viscometer constant (K), which contains all of the information about the dimensions of the viscometer tube and ball. 2 . . (21) In eq 21, g is the acceleration of gravity in cgs units, r is the ball radius in cm, d is the ball diameter in cm, D is the tube diameter in cm, and L is the distance of the ball-drop in cm. Equation 22 is used to calculate the viscosity (η). (22) In eq 22, ρb is the ball density in g/mL, ρs is the sample density in g/mL, and t is the time of the ball-drop in seconds. Wetting Index The wetting index (W) is a measure of how well a solvent will physically be able to wet a surface and flow under particulates on the surface. This wetting index is focused on the properties of the solvent, and not the interactions that the solvent has with any particular surface. In precision cleaning, it is good for the solvent to have a good solvency for the solute of interest, but it is better for the solvent to also be able to wet the entire surface of the item that is being cleaned. The entire surface of the item includes grooves, holes, threads, and any other areas that may be soiled. A higher wetting index indicates that the solvent will wet the surface better, while a lower wetting index indicates the opposite. The wetting index relies on density in g/mL, surface tension in mN/m, and viscosity in cP. This was the main reason for measuring these properties in this study. The equation for calculating the wetting index was deduced from the data in the Kanegsberg book, and is shown here in eq 23.18 (23) A low surface tension and viscosity allows the solvent to access grooves, small holes, and threads. A large density allows the solvent to impact the surface with more force when a spray or ultrasonic bath of the solvent is used for cleaning.