Solvent Splitter

The solvent splitter allows the manufacturer to separate the solvent mixture back into MEK and MIBK or Toluene which can then be added back to the solvent mixture in the plant to optimize the solvent composition and minimize the pour - filter temperature spread to achieve maximum throughput. 

---

The ideal would be to operate as close to the miscibility curve as possible. However, this curve shifts depending on the Basestock. Wax molecules in heavier grades come out of solution earlier and this has the effect of shifting the miscibility curve to the left. Plants equipped with solvent splitters to separate the MEK from the MIBK or Toluene after it has been recovered in the DWO and Wax recovery sections may blend to the optimum solvent composition for each Basestock. Manufacturers without the capability to change solvent composition will set the plant solvent composition based on... 

Figure 47. Typical Solvent Splitter Flow Diagram...

Although current electrospray interfaces can be used with flow rates of up to 1 mL min-i, lower flow rates, e.g. 40-200 iL min i, are preferred in most LC-MS applications, thus requiring the use of either a postcolunm solvent splitter or microbore LC colunrn. 

Morristown, NJ) for the ion source. No carrier gas separator was used. For determination of nitrosamines and TBDMS derivatives of hydroxy-nitrosamines, columns and operating conditions were identical to those for GC-TEA analyses For most work, the He flow rate was 15 cc/min and the column effluent was split 1 1 between a flame ionization detector and the mass spectrometer. The stainless steel splitter, solvent vent valve (Carle Instruments, Fullerton, CA), and associated plumbing were... 

Maximum flow-rates compatible with DLI interfaces are in the range of 50 to 100 pL/min. Microbore columns (<1.0mm i.d. column) operating at 5 to lOOpL/min are ideally suited for DLI LC/MS. A flow splitter is required to couple conventional LC with a DLI interface so that only a fraction of the total eluent is introduced into the mass spectrometer. Splitting the flow outside the mass spectrometer results in loss of sensitivity, an undesirable consequence. Henion and co-workers have reported an approach in which the splitter is incorporated into the desolvation chamber of the mass spectrometer. The removal of solvent is achieved by diverting the vapor generated by the solvent without loss of sample and, therefore, sensitivity. Excess pressure inside the mass spectrometer is therefore avoided while higher flow-rates can be accommodated. 

The product coming out of the reactor consists of excess hydrogen and a reformate rich in aromatics. Typically the dehydrogenation of naphthenes approaches 100%. From 0% to 70% of the paraffins are dehydrocyclized. The liquid product from the separator goes to a stabilizer where light hydrocarbons are removed and sent to a debutanizer. The debutanized platformate is then sent to a splitter where Cg and C9 aromatics are removed. The platformate splitter overhead, consisting of benzene, toluene, and nonaromatics, is then solvent extracted (46). 

A reversed-phase HPLC Cl8 or C30 narrow-bore column is typically used for LC/MS with APCI. Details about chromatography columns used for carotenoids are contained in unit F2.3. For most APCI systems, the optimum flow rate into a mass spectrometer or tandem mass spectrometer equipped with APCI, as controlled by a syringe pump or HPLC pump, is usually between 100 and 300 pl/min, which is ideal for narrow-bore HPLC columns. Larger diameter columns should be used with a flow splitter postcolumn to reduce the solvent flow into the mass spectrometer. For example, if a 4.6 mm i.d. column was used at a flow rate of 1.0 ml/min, then the flow must be split postcolumn 5 1 so that only 200 pl/min enters the mass spectrometer. 

An HPLC chromatograph was used together with a mass detector (when required, a stream splitter (app. 10 1) was inserted between the column and the detector). A column (250 X 4.6-mm ID) of NUCLEOSIL 5SA was flushed with 1% aqueous ammonium nitrate solution at a flow rate of 0.5 ml/min for 1 h, then with distilled water at 1 ml/min for 1 h. Silver nitrate (0.2 g) in water (1 ml) was injected onto the column in 50-/zl aliquots at 1-min intervals silver began to elute from the column after about 10 min. Twenty minutes after the last injection, the column was washed with methanol for 1 h, then with 1,2-dichloroethane-dichloromethane (1 1) for another hour. The three solvent reservoirs contained the following (A) 1,2-dichloroethane-dichloromethane (1 1) (B) acetone and (C) acetone-acetonitrile (9 1). For linoleic acid-rich seed oils, gradients of A were employed to 50% A-50% B over 15 min, then to 50% B-50% C over a further 25 min and held there for 5 min. For linolenic acid-rich seed oils, C was changed to acetone-acetonitrile (4 1), and the flow rate was increased to 1 ml/min gradients of A were utilized to 50% A-50% B over 10 min, then to 70% B-30% C over 20 more min, and finally to 100% C over another 30 min. 

B. Pacciarelli, B. Muller, et al., GC column effluent splitter for problematic solvents introduced in large volumes determination of di-(2-ethylhexyl)phthalate in triglyceride matrices as an application, HRC CC J. High Res. Chromatogr. Chromatogr. Commun., 11 135-139(1988). 

Sample introduction is a major hardware problem for SFC. The sample solvent composition and the injection pressure and temperature can all affect sample introduction. The high solute diffusion and lower viscosity which favor supercritical fluids over liquid mobile phases can cause problems in injection. Back-diffusion can occur, causing broad solvent peaks and poor solute peak shape. There can also be a complex phase behavior as well as a solubility phenomenon taking place due to the fact that one may have combinations of supercritical fluid (neat or mixed with sample solvent), a subcritical liquified gas, sample solvents, and solute present simultaneously in the injector and column head [2]. All of these can contribute individually to reproducibility problems in SFC. Both dynamic and timed split modes are used for sample introduction in capillary SFC. Dynamic split injectors have a microvalve and splitter assembly. The amount of injection is based on the size of a fused silica restrictor. In the timed split mode, the SFC column is directly connected to the injection valve. Highspeed pneumatics and electronics are used along with a standard injection valve and actuator. Rapid actuation of the valve from the load to the inject position and back occurs in milliseconds. In this mode, one can program the time of injection on a computer and thus control the amount of injection. In packed-column SFC, an injector similar to HPLC is used and whole loop is injected on the column. The valve is switched either manually or automatically through a remote injector port. The injection is done under pressure. 

The upper position allowed isocratic and gradient micro-LC with the possibility to assist the separation with an electric field. The capillary was introduced into the electrode and the solvent was delivered to the capillary via the inner channel. The solvent overflow was liberated through the outlet capillary, which had a higher flow resistance in this case. In this position, the special vial functions as a flow splitter. 

Ozone Competition Reaction Procedures. In the relative rate studies, the solvent-saturated ozone—oxygen stream was passed into a glass bubbler reactor vessel charged with 4 ml of about 4 X 10"2M concentration of each of the two silanes to be competitively ozonized as well as of an inert saturated aliphatic hydrocarbon to function as internal standard. (For example, n-undecane was used for the tributylsilane/trihexylsilane study.) The effluent from the reactor passed through a solvent-filled bubble counter to visualize the flow. The inlet stream splitter mentioned earlier was adjusted to allow 2—4 hours for each runs completion, as determined by experience. The temperature was controlled at 0°C in both the saturator and the reactor by ah ice bath. 

---