Fractionator minimum stages

For multicomponent systems, an approximate value of (he minimum number of stages (at total reflux) may be obtained from the Fanske relationship [Eq. (5.3-28)]. In the use of this relationship for multicomponent mixtures, die mole fractions and die relative volatility refer to die light and heavy keys only. However, values for the nonkey components may be inserted in the equation to determine (heir distribution after the numbet of minimum stages has been determined through the use of the key components. For a more rigorous approach to the determination of minimum stages, see die paper by Chien/... 

The Fenske and Winn equations are not restricted to the two key components. Once JVmin is known, they can be used to calculate mole fractions and Xi for all nonkey components. These values provide a first approximation to the actual product distribution when more than the minimum stages are employed. A knowledge of the distribution of nonkey components is also necessary when applying the method of Winn because (12-15) cannot be converted to a mole ratio form like (12-12). 

Find minimum stages, minimum reflux, and ideal stages. The vapor liquid data are given below (mole fractions) ... 

A two-colunm system is used to produce anhydrous (absolute alcohol). The aqueous alcohol feed (89 percent alcohol 11 percent water) is separated in column 1 to the anhydrous product and an overhead which is condensed into a benzene and water layers. The benzene layer is used as a reflux to column 1. The water layer is the feed for column 2 (bottom product is water). Its top product is recycled to column I s condenser. The bottom mole fractions for column 1 are 0.999 for alcohol, 0.0009 for water, and 0.0001 for benzene. Reflux to column 1 has the following mole fractions (0.217 alcohol 0.256 water 0.527 benzene). The relative volatilities for benzene, alcohol, and water are 3.60, 0.89, and 1.0 at the column bottom and 0.62, 0.47, and 1.00 at the reflux. Estimate minimum stages required. 

LK = subscript for light key Nn, = minimum theoretical stages at total reflux Xhk = mol fraction of heavy key component Xlk = mol fraction of the light key component otLK/HK = relative volatility of component vs the heavy key component... 

N,n = Minimum theoretical stages at total reflux Q = Heat transferred, Btu/hr U - Overall heat transfer coefficient, Btu/hrfP"F u = Vapor velocity, ft/sec U d = Velocity under downcomer, ft/sec VD(js = Downcomer design velocity, GPM/fL Vioad = Column vapor load factor W = Condensate rate, Ibs/hr Xhk = Mol fraction of heavy key component Xlk = Mol fraction of the light key component a, = Relative volatility of component i versus the heavy key component... 

Gilliland" tells how his famous correlation was developed for relating actual and minimum reflux to actual and minimum theoretical stages for a fractionating column. Numerous plate-to-plate calculations were made and the results plotted using his well-known correlating parameters. The best curve was then drawn through points. 

Consider fibers that all have the same strength and are relatively brittle in comparison to the matrix as studied by Kelly and Davies [3-26]. Moreover, both the fibers and matrix are active only in the linear elastic range (stage 1 in Figure 3-46). If the composite material has more than a certain minimum volume fraction of fibers, V, the ultimate strength is achieved when the fibers are strained to correspond to their maximum (ultimate) stress. That is, in terms of strains. 

Van Driesen and Stewart (V4) have reported temperature measurements for various locations in commercial gas-liquid fluidized reactors for the large-scale catalytic desulfurization and hydrocracking of heavy petroleum fractions (2500 barrels per day capacity). The hydrogenation was carried out in two stages the maximum and minimum temperatures measured were 774° and 778°F for the first stage and 768° and 770°F for the second. These results indicate that gas-liquid fluidized reactors are characterized by a high effective thermal conductivity. 

A typical grade efficiency curve for the product classification step is given in Figure 4. A value of nearly 100 percent is attained at large sizes, whereas normally a value equal to or larger than the so-called dead flux is attained at small sizes. This is caused by the diluted discharge of the coarse fraction. It represents the minimum amount of residual fines in the product after one separation stage. 

Mixture designs are applied in cases where the levels of individual components in a formulation require optimization, but where the system is constrained by a maximum value for the overall formulation. In other words, a mixture design is often considered at this stage when the quantities of the factors must add to a fixed total. In a mixture experiment, the factors are proportions of different components of a blend. Mixture designs allow for the specification of constraints on each of the factors, such as a maximum and/or minimum value for each component, as well as for the sum and/or ratio of two or more of the factors. These designs are very specific in nature and are tied to the specific constraints that are unique to the particular formulation. However, as with the discussion of the fractional factorial designs, in order to be most efficient, it is important to provide realistic prior expectations on anticipated effects so the smallest design can be set up to fit the simplest realistic model to the data. 

Like other early stage R D alliances, a large fraction of total pre-commercial payments from pharma licensee to biotech licensor in CC alliances comes in the form of R D reimbursement. Such reimbursement is usually negotiated as a minimum annual payment per year (R D/yr), either on a fixed dollar basis or as a function of the number of biotech full-time equivalent scientists (FTEs) expected to conduct the alliance program. In addition, R D reimbursement is usually negotiated for an expected number of years (R D term), although often it is possible for the licensee to cancel R D reimbursement prior to the completion of the R D term. 

For standard types of finite-stage contactor columns operated in the range of allowable velocities where the overall column efficiencies are essentially constant, O Connell has correlated efficiency data on the basis of liquid viscosity and relative volatility (or gas solubility). The results for fractionators and absorbers are presented in Fig. 16-9. This correlation is based, primarily, on experimental data obtained with bubble-cap columns having a liquid path of less than 5 ft and operated at a reflux ratio near the minimum value. Figure 16-9 is adequate for design estimates with most types of commercial equipment and... 

---