Dead water

Fic. 7. Properties of the F, C, I, E curves for particular patterns of flow (dead-water regions and bjrpassing flow absent) in closed vessels (L13). 

As a model of a certain poorly agitated continuous stirred-tank reactor of total volume V, it is supposed that only a fraction w is well mixed, the remainder being a dead-water zone. Furthermore, a fraction / of the total volume flowrate v fed to the tank by-passes the well-mixed zone completely (Fig. 2.24a). 

There is much work to do to recover from the misestimate that entered into calculations made after Debye and HUckel by neglecting ion pairing and dead water (Section 3.8). 

No great superiority of one approximation over the other is apparent. However, tha-e is something a little inconsistent, even dubious, about the comparisons. The approximation marked Bjerrum allows for ion association, and this is a fact known to be present at higher concentrations, as independently shown by Raman spectra of electrolytes containing, e.g., NOj ions. The other models assume, however, that all the ions are free and unassociated, even in concentrations in which experiment shows strong ion-pair formation. Also, where is the allowance for Bjerrum s dead water, ... 

Suppose we denote the liquid in dead-water regions as having compositon C, and liquid within the main bulk flow to have composition C. The liquid hold-up in the two regions would in general be different, but the sum of the two would be... 

WATER — Dead Water. The Water of Sulphur which is known to the Philosophers. 

M. Sharpe, Deadly Waters Run Deep The Global Arsenic Crisis, J. Envir. Monitoring, 5(5), 81-85N (2003). 

Several models have been suggested to simulate the behavior inside a reactor [53, 71, 72]. Accordingly, homogeneous flow models, which are the subject of this chapter, may be classified into (1) velocity profile model, for a reactor whose velocity profile is rather simple and describable by some mathematical expression, (2) dispersion model, which draws analogy between mixing and diffusion processes, and (3) compartmental model, which consists of a series of perfectly-mixed reactors, plug-flow reactors, dead water elements as well as recycle streams, by pass and cross flow etc., in order to describe a non-ideal flow reactor. 

Fig.4.3-3 demonstrates two perfectly-mixed reactors. Reactor 1 contains a "dead water" element of volume Vd [21, p.296] where the other part of volume Vi is perfectly mixed. The total volume of the reactor is Vi-i-Vd and the volume of the second reactor is V2. A tracer in a form of a pulse input is introduced into reactor 1 and is transferred by the flow Qi into reactor 2 where it is accumulating. 

Fig.4.3-3. Perfectly-mixed reactors with a "dead water" element in...

Fig.4.4-2. Plug flow reactor with a "dead water" element...

A new method for measuring environmental tritium levels without enrichment was recently developed by the authors (Cawley t al , 1980). This method is basically a double isotope, external standard count procedure which eliminates the background by measuring tritium free water scimples ("dead water" samples) and applying a difference method. 

The most delicate part of the method developed to measure environmental tritium levels is the elimination of the background. A difference method has been applied for this purpose using "dead water" samples, i.e. tritium free water samples obtained from National Bureau of Standards. 

The environmental and dead water samples exhibit quenching characteristics falling into the range of 0.5 - 0.6 quenched standards (Packard Instrument Co.) for tritium and carbon-14. During each cycle an efficiency calibration using the quenched standards is made. A cycle consists of counting the 15 sample vials, 2 or more vials of dead water, and the 4 quenched standards (two carbon-14 and two tritium standards). The selected count period is 20 minutes. 

C = efficiency for counting carbon-14 in Channel II in dead water-... 

The errors associated with estimating the efficiencies in each channel are much smaller than the errors associated with the count rates in each channel. Therefore, the errors of the efficiencies are neglected. It is assumed that the tritium content of the dead water is zero (National Bureau of Standards certifies that it is less than. 6 pCi/1). 

The numerous short counting periods, the large number of vials coupled with the efficiency calibration data from the external standard count, the quenched standards, and the vials with mixtures of either scintillator-sample or scintillator-dead water make this a program of data organization and reduction. To eliminate the need to change the source progreun due to different numbers of observations from one sample run to another, the PL/1 feature of dynamic allocation of storage has been used. This allows the user to define the dimensions of the arrays for the input data. It is not uncommon for the arrays for the quenched standards, the dead water, and the sample data to have different dimensions. 

The data from the quenched standards and the AES for the sample and dead water are used to calculate the efficiency of... 

Through the external standard counts of the quenched standards and those of the dead water and of the sanple, the counting efficiencies are calculated by subroutine calls. 

Once the efficiencies for the sample and the dead water and their respective count rates in Channels I and II are determined, the tritium concentrations in pico-Curies per liter and the associated error may be calculated for each sample vial. 

The presence of separated regions in the flow and stagnant areas cause the particles to settle in dead water zones and it becomes very difficult to resuspend them. 

By inserting 50 mm packings in a 1 m diameter column the Pe-number is decreased. This may be caused by statistical nonuniformities of the packing. Dye added to the water input shows the formation of channels other parts of the liquid remain much longer in dead water regions. The resulting extension of the residence time curve yields the extremely low Pe-number. 

---