Development of I
- --
Chromatography Countercurrent chromatography (CCC) designates a family of chromatographic methods with the common feature of not employing a solid support (Z-3).Thus, CCC offers advantages over other chromatographic methods in that it is free from all complications arising from the use of solid supports. Adsorptive loss and denaturation of samples, contamination, and tailing of solute peaks are minimized. The development of CCC necessitated an entirely novel approach to achieve a high partition efficiency which is readilv available with solid supports. The followine were some of the difficult problemsencountered The stationary phase had to be securely retained in the open space of the column through which the mobile phase is continuously eluted. The column space had to be divided into a number of partition units to produce hundreds or thousands of theoretical plates. Mass transfer resistance had to be minimized by providing broad interface area and efficient mixing between the two phases without introducing undesirable sample band broadening effects. Surprisingly enough, all of these problems have been solved simply hy using a coil as the separation column. 534A
Origin of CCC About 20 years ago the development of CCC was initiated by a series of experiments on the hydrodynamic behavior of solvent phases in a rotating coil. Figure 1schematically illustrates the motion of equal volumes of two immiscible solvents confined in a slowly rotating coil. The coil at the top contains the lighter phase (white) in the left half and the heavier phase (blue) in the right half, with the interface at the bottom of the coil. When the coil is rotated slowly around its own axis, the two phases undergo countercurrent movement across the original interface. For every half turn of the coil a new interface is created at the bottom of the coil, and finally the coil is filled with alternating segments of the two phases. This countercurrent interaction of the two phases suggests that solutes, if introduced beforehand at the original interface, would be distributed along the length of the coil according to their partition coefficients. This idea was successfully tested with the coil planet centrifuge (4,5),which was designed to facilitate countercurrent movement of the two phases along the narrow bore of the coil by introducing slow rotation of the coiled column in a strong centrifugal force field. Theca-
ANALYTICAL CHEMISTRY, VOL. 56, NO. 4. APRIL 1984
pability of the method was first demonstrated by the separation of pigments in a long coiled column of narrow-bore tubing as shown in Figure 2. The coil was first half-filled with the upper phase in the left half, and the sample (segregated by air bubbles which later dissolved) was introduced from the right side, followed by the lower phase. Then, coil planet centrifugation yielded an efficient separation of four components as shown in the top coil. The remaining coils show the reproducibility for each component. This original CCC scheme using endclosed coils has been radically improved by introduction of a flowthrough mechanism which facilitates continuous elution, monitoring, and fraction collection as in LC.
Two Basic CCC Systems The principles of two basic continuous elution CCC systems are illustrated in Figure 3. One is called the hydrostatic equilibrium system (HSES) and the other the hydrodynamic equilibrium system (HDES). The hydrostatic system uses a stationary coiled tube. The coil is first filled with one phase of an equilibrated two-phase solvent system. The other phase, introduced at one end of the coil, percolates through the first 0003-2700/84/035 1-534A$01.5010 0 1984 American Chemical Society
CE----t Yoichiro It0 Laboratwy of Technical Development National Heart.Lung, and Blood Institute Bethesda. Md. 20205
Waiter D. Conway Department of Pharmaceutics
School of Pharmacy State University of New Ywk at Buffalo Amherst, N.Y. 14260
phase on the front side of the coil in Figure 3, coalesces, and ascends the coil to the next turn where the process is repeated, leaving stationary segments of the first phase in each coil unit. The hydrodynamic system uses a similar setup except that the coil is slowly rotated around its own axis, thereby rotating relative to the gravi. tational field. Introduction of this simple motion has amazingly complex effects on the beliavior of the two phases in the coil. Rotation makes the coil asymmetric. Particles or droplets introduced into the coil move toward one end of the coil. This end of the coil is called the head and the other end is the tail. A rotating open coil is an Archimedean screw which pumps the liquid within it from the tail to the head end. When the mobile phase is introduced through the head of the coil, displacement of the stationary phase is opposed by the Archimedean pumping action with the result that a hydrodynamic equilibrium is established in which each phase occupies nearly 50%of the space in each coil unit. In addition, rotation causes constant mixing of the two phases. In both the hydrostatic and the hydrodynamic systems, after the mobile phase reaches the tail of the coil it dis-
places only itself, leaving the stationary phase uniformly distributed along the column as segments of constant volume in each turn of the coil. Solutes introduced at the mobile phase inlet are subjected to a partition process between the two phases and are separated according to their partition coefficients. Each system has specific advantages. The hydrostatic system offers simplicity and stable retention of the stationary phase at slower flow rates, whereas the hydrodynamic system offers more efficient mixing of the two phases, achieves higher partition efficiency, and permits a higher mobile phase flow rate. In the hydrostatic system, one side of the coil is entirely occupied by the mobile phase and hecomes dead column space, while in the hydrodynamic system efficient mixing takes place in every portion of the coil. The performances of these two hasic CCC systems have been compared by separation of dinitrophenyl (DNP) amino acid samples in a two-phase solvent system composed of chloroform, acetic acid, and 0.1 N hydrochloric acid ( 2 2 1 ) using a simple rotary device eauiDDed with a Dair of rotary seals. The r&ults clearly indicate that the hydrodynamic system yields much greater resolution.
Figure 1. Countercurrent process of the two immiscible solvents confined in a rotating coiled tube
ANALYTICAL CHEMISTRY, VOL. 56. NO.
4, APRIL 1984
535A
ydrwtatic Equilibrium System
U
Hydrodynamic Equilibrium System
L 4
7
Figure 2. Separation of pigments in a coiled tube by coil planet centrifugation MQ. nmihylween: MB, memylene blus: NR, new Ira1 rsd: FB. basic I ~ s I n e
Development of Efflcient CCC Schemes Based on the HSES The basic hydrostatic system has heen improved in two different directions to develop both preparative and analytical schemes, as illustrated in Figure 4. For preparative applications, the inefficient portion of the coil entirely occupied by the mobile phase was reduced to a narrow transfer tube while the effective portion of the coil was replaced by a straight tubular column. In droplet CCC (6.7)the mobile phase introduced into the vertical column forma multiple droplets in the stationary phase to divide the column space into numerous partition units. Although the scheme is simple and efficient, the operation is time-consuming and the necessity of droplet formation Limits the choice of solvent systems. An alternative scheme employed a locular column made by placing centrally perforated disks into the column at regular intervals. In rotation locular CCC (6), the retention of the stationary phase and the interfacialivea between the two phases are optimized by inclination of the column, while mixing of the two phases is accomplished by rotation of the column. This scheme permits universal application of conventional solvent systems. In gyration locular CCC (6).the locular column is held vertically and gyrated to induce circular movement of the two phases and their interface within each compartment. This gyrating motion eliminates the need for rotating seals, which are a potential source of leakage and contamination. Reduction of both the internal diameter and the helical diameter of the coil and application of a strong centrifugal force field to the column produced an improvement in the inferior performance of the hasic hydrostatic system, making it suitable for analytical applications where small sample volumes are employed (Figure 4). The 538A
ANALYTICAL
Figure 9. Two basic CCC systems Repinted from RBterence 1
increased hydrostatic pressure resulting from the centrifugal field could be substantially reduced by using a column employing a twisted pair of tubing strands (6).
Rotary-Seal-Free Flow-Through Centrifuge Schemes The performance of both basic CCC systems was further improved by applying various patterns of centrifugal force fields. The centrifugal apparatus developed for performing CCC is equipped with a rotary-seal-free flow. through mechanism which permits continuous elution through multiple flow channels without risk of leakage or contamination, even under a high pressure of several hundred psi. Consequently, these systems provide a re. liable elution system comparable to other chromatographic methods and are broadly applicable to the separa-
tion and purification of various biological materials. The principles of various types of rotary-seal-free flow-through centrifuge schemes and their mutual relationships are illustrated in Figure 5. Each diagram indicates the orientation and motion of the cylindrical column holder from which emerges a bundle of flow tubes, the ends of which are tightly supported at a point on the central axis. The planetary motion of the holder, indicated by arrows, is described as rotation (about its own axis) and revolution (around the central axis of the centrifuge). These schemes are divided into three classes-synchronous, nonplanetary, and nonsynchrono-ccording to their modes of column motion, and are numbered I through VI1 in the order of their complexity. In the synchronous series, rotation and revolution of the holder are synchronized so
Figure 4. Development of efficientCCC schemes from the basic hydrostatic equilibrium system (HSES) Reprinted Iran Reterenoe 1
CKMiSlRY. VOL. 56, NO.
4. APRIL 1984
--
d a dyour fill of sarnw-handling bykmd?
me the Gikon Zl2 Liquid Handler with Multi-level Interfacing
1
- lulti-level Interfacingis
a concept developed by lson to make the 212 Liquid Handler product specific for your liquid handling application. A shown in the diagrams the 212 may be used as a stand-alone device or in a variety of other configurations through the extensive input-output capabilities accessible through ~ . i e212.
I
i e r e are some of the benefitsyou'll realize by using the 212: High throughput-large liquid handling capability combined with the potential for 24 hour operation.
-
-
I--
Unattended operation increases productivity by freeing personnel for other work. Meaningful data assured by eliminatingtechnician error on routine repetitivetechniques. Simple operation using promptlrespond format.
--
Easy to read vacuum fluorexent display
r Ih
Economically priced; reduces labor:ensive tasks
Durable construction assures long operating life for continued value.
1
For more information, call our 212 product specialist collect at 16081836-1551.
1 Circle
@ tor literatute.
Circle
Box 2i: 3000 W Belfline Middleton, WI 53562 16081836-1551 Telex: 265478
72 rue Gambetta. B.P Na 45,95400 Villiers-le-Bel. FRANCE Tel. (3199054.41 Telex: 696682
GILSON TERNATIONAL
0 #08
have y w representative all on us.
ANALYTICAL CHEMISTRY,
VOL.
56. NO. 4, APRIL 1984
0
537).
-
-
laure - 5. Flow-throuah centrifuae schemes free of rotary seals Reprinted from Relerence 1
that one revolution of the holder about the central axis gives zero (I), one (11,111). or two (IV) rotations of the holder with respect to an outside observer. In the nonsynchronous series, the rates of revolution and rotation are independently adjusted. In the nonplanetary series, rotation and revolution share a common axis resulting in either no rotation or a single rotation (V) around the central axis of the centrifuge. In scheme I, the holder counterrotates about its own axis while revolving around the central axis. As a result, the revolving holder always keeps the same orientation with respect to the outside observer. Because the counterrotation of the holder constantly unwinds the twist in the tube bundle made by the revolution, the tube bundle does not kink. This principle applies also to the other synchronous schemes with tilted (II), horizontal (III), and even inverted (IV) orientations of the holder. When the holder of scheme I is brought to the central axis of the centrifuge, the counterrotation of the holder cancels out the revolutional motion, resulting in no rotation of the holder. However, a similar shift applied to the holder of scheme IV yields quite an unexpected result. Because 538 A
-
rotation and revolution of the inverted holder in scheme IV are both in the same direction, the central shift of the holder results in the addition of these two motions; hence the holder rotates at doubled speed. This shift also requires revolution of the flow tubes around the central axis. Scheme V provides the basis for the nonsynchronous schemes. Moving the holder away from the central axis to a position analogous to schemes I or IV provides schemes in which the revolutional rate of the column holder may be freely adjusted independent of the rate of revolution of the holder about the central axis. Analysis of the Synchronous Planetary Motion Each scheme in Figure 5 produces a
characteristic centrifugal force field. The nonplanetary scheme, scheme V, produces a stable radial centrifugal force field as in a conventional centrifuge, whereas schemes I1 and I11 display the most complex patterns, which involve three-dimensional fluctuations of the centrifugal force vectors. In the rest of the schemes, the force vectors are always confined in a plane so that the mathematical analysis can be conveniently performed using an x-y coordinate system. Simple mathemat-
ANALYTICAL CHEMISTRY, VOL. 56, N O . 4. APRIL 1984
Princeton instruments, /nc Examines the principles and applicationsof optoelectronic image detectors (OIDs) as multichannel spectrometric devices. Expands on Symposium Series No. 102, "Multichannel Image Detectors," including VUV-X-ray and electron spectroscopy as well as UV-NIR spectral regions. Deals with theoretical and fundamental aspects of multichannel image detectors and compares their performances.
-
-
CONTENTS -. .._... Guidelines lor Selenion of Four OptoBIecIronic Image Detectors for Low-Light Level Applications Silicon-lnlensifiedTarget Vidicon Detector: Operation. Charaneriratian, and Application in Atomic Spectroscopy Resarch Wide Dynamic Range Array Detector tor Absorbance and Rotation Spectrometry InductivelyCoupled PiasmaAtomic Emission Spectrompy with Multichannel Array Detection Multielement Emission Speclromelry Using a Chargelniection Device Detector Charae-lniection
*
*
I Balloon,and Spacecratl Experiments
Order horn: Arneriun Chenlul s~clely DI8hlbYtiOn M c e Dept. 55
iis~si~tmrnsi., N.W.
W"hlnpton, DC zw% or CALL TOLL FREE 8004244747 and use y w r VISA or U ~ b I C a r d .
I
The VISTA 402 LC/GC Data System handles all thesesamples fully -
._r --*T-
I
i -
,+--
,: ,
With a VISTA 402 Data System controlling a Varian VISTA LC and GC,your repertoire of LC or Gc analyses is nearly unlimited. Using the unique intelligence of a single VISTA 402, all the samples shown can be analyzed using a VISTA LC and GC.And you can instruct the system to orchestrate this imDres. sive arrav of analvses in vour absence. VISTA 402-initiated LC and GC automation is much more than just repetitive injectis, more than merely adding an autosamPiing device. The kev is our conceDt of "user emulation:' a q & c h that strives to simb.ate the actims you woulo perform .iycu were running the system mawally A VISTA system will choose chromatographyldata han.
halocarbons m drinking w a t ~ r
dling parameters based on the identity of the sample. It will perform a prep run for equilibration, control all GC andlor LC parameters.check on resolution. and optimize data processing. It will "look at the peaks as they are eluted and control signall noise, tangent skim, horizontal or valley baselines, and other data handling parameters, It will perform all resuit caiculations-calibrations and analyses, area or height calculation, auto calibration, and check for instrument, resolution. and calibration errors. The VISTA system will plot the chromatogram with or without baselines and print the results, including relative retention time, peak widths at half height, and a run log and error logs. Data and results storage are automatic, of course! In a nutshell: A VISTA 402-contr01led LC or GC emulates all your actions and decisions as closely as non-humanly possible. That's what we mean by user emulation! And here's the best part: the VISTA LClGC shown above is only working at half its capacity! The 4-channelVISTA 402 can still handle two more LCs or Gcs with autosamplers, plus data system options-which means you can double your laboratory's throughput whenever the time is right.
varian
/nte//igenfChromatography..
.from Varian
Scheme I
Scheme IV
Center of Revolution
1Centerof Rotation
Center of RotatirP
.e,
'
X
Orbits
ical analyses have been carried out on the synchronous planetary motions of schemes I and IV; the results are summarized in Figure 6. For schemes I and IV, (a) shows the planetary motion, (b) the coordinate system for analysis, (c) the orbit of the arbitrary point on the holder, and (d) the distribution of the centrifugal force vectors at a given moment. The location of the arbitrary points on the holder is expressed by IS = r/R, where r denotes the radius of rotation (the distance between the arbitrary point, P, and the holder axis, Q ) and R, the radius of revolution (the distance between the axis of the centrifuge, 0, and the holder axis, Q). Scheme I produces a circular orbit regardless of the location of the point on the holder. Every point on the holder is therefore subjected to a homogeneous centrifugal force field. Each centrifugal force vector uniformly rotates around the point to produce efficient mixing of the two phases in the coil as in the basic hydrodynamic system. This scheme is particularly suitable for analytical separations with a narrow-bore coil. Scheme IV, on the other hand, yields a complex orbit which changes its shape from a circle (j3 < 0.25), to a heart (j3 = 0.5). to a double loop (@= 1.0) as the location of the point becomes more distant from the axis of the holder. This gives a highly complex heterogeneous distribution of the centrifugal force vectors. However, when the location of the point is some distance away from the holder axis (j3> 0.25), the force vectors are always directed outward from the holder. This unique distribution of the force vectors produces different hydrodynamic equilibrium patterns of the two phases according to the location and orientation of the coiled column on the holder. When the coiled column is eccentrically mounted on the holder, the outward-directed force vector separates the two phases in such a way that the heavier phase occupies the outer portion and the lighter phase occupies the inner portion of each helical turn as in the basic hydrostatic system. However, vigorous oscillation of the force vector, as well as rotation of the holder relative to the centrifugal field, produces an efficient mixing of the two phases in the coil, which improves the partition efficiency. This coil orientation produces excellent resolution in both preparative and analytical ranges. When the coiled column is mounted coaxially around the holder, the hydrodynamic distribution of the two phases is completely altered to produce a novel CCC scheme called high-speed CCC, which will be described later in detail.
I
Figure 6. Mathematical analyses of two synchronous planetary motions Reprinted hMn Reference 1
540A
ANALYTICAL CHEMISTRY, VOL. 56, NO. 4. APRIL 1984
Forget coulometric versus volumetric.
At last, Karl Fischer titrations over the full range...lppm to 100%. Now, just one microprocessorcontrolled instrument gives you the entire range of water content determinations. It's the Mettler DL18 Karl Fischer Titrator. It has the range of two instruments, but the price tag of only one...under $4,000.
Fasf, easy,accurate. Karl Fischer titration procedures are routine tasks performed daily in all types of laboratories,from food and chemicals to plastics and petroleum. Gaining precise information about water content in samples can be a chore. The DL18 turns this into an easily accomplished task, all at the push of a button. Automatic results
in mg, percent, or ppm of water are quickly and reliably determined.Very simple and extremely accurate. A complete, clean system. You don't need expensive accessories when you purchasethe DLl8. Everything you need is included, even an automatic device for changing the tirant (all conventional and pyridine-free tirants may be used),and the computer for result calculations. Incidental items, such as titration beakers and drying media are also standard equipment.
A complete Ilne. If you perform other types of titrations in your lab, take a look at the CIRCLE 147 ON READER SERVICE CARD
New.The Mettler DLl8 automatic Karl FiScherTitrator.
Mettler DL40RC
[email protected] all-purpose instrument provides extremely accurate information-at the touch of a button, as well. Its versatility and ease of operation are second to none. Write or phone today for more information, or a demonstration. Mettler Instrument Corporation, Box 71, Hightstown, NJ 08520. Phone (609) 448-3000.
I Figure 7. Design principle through coil plarmt centrifuge
Flgure 8. Prototyps of the combined horizontal flow-through
low-
coil planet centrifuge Reprhted horn Refereme 10
R a p r i m hom Anal. Bkhem. 1979, loo. 271
Examples of Centrifugal CCC Devices Free of Rotary Seals Combined horizontal flowthough coil planet centrifuge (8-10). The principle of this apparatus is shown in Figure 7.Diagrams labeled I through IV are identical to those in the synchronous schemes shown in Figure 5. Schemes I and IV for preventing twisting of the flow tubes are paired in one apparatus. The pulley-driven holder produces scheme I synchronous planetary motion suitable for analytical separations, and the gear-driven holder produces scheme IV synchronous planetary motion suitable for preparative separations. Figure 8 shows our prototype of the combined horizontal flow-through coil planet centrifuge. The analytical
Galumn 8m i.d.. 24-mL Capaciw, Pulley Side Solvent System: CHCI,:CH,COOH:0.1N HCi (221) Slatbnary Phase: Nonaqueous Phase Sample: DNP Amino Acids Sample Volume: 50 plL Rmiulion: SW rpm DNP-Asp Flow Rate: 6 mUh
DNP-Giu
DNP-Val
DNP-Pro
based on the nonplanetary scheme V in Figure 5, is illustrated in Figure 10. The center disk rotates around the central axis of the centrifuge at an angular velocity of 2 o,and the tube bundle revolves around thesame axis at half that speed, w. The scheme produces a stable radial centrifugal force field as in the conventional centrifuge system and allows flow in and out through the rotating disk without the use of rotary seals. The design of the apparatus is shown in Figures l l a and b. In the cross-sectional view through the central axis of the apparatus (a), the rotary frame holds three rotary elements, the countershaft (right), the central disk, which is in reality a shallow bowl (center), and the tube-supporting hollow shaft (left). The coupling of the lower pulley of the coun-
column mounted on the pulley-driven holder (upper) has 3500 helical turns of 0.55-mm i.d. PTFE tubing with a total capacity of 24 mL. The preparative column mounted on the gear-driven holder (lower) has loo0 helical turns of 2.6-mm i.d. PTFE tubing with a total capacity of about 270 mL. DNP amino acid separations obtained from these columns are shown in Figures 9a and b. Several other CCC devices based on the synchronous schemes (Figure 7) include the angle rotor CCC (11) ( I I ) , the elution centrifuge (111) (1% and the toroidal coil planet centrifuge (IV) (13).They are all capable of producing highly efficient analytical-scale separations. Toroidal coil centrifuge ( 1 4 , l S ) . The principle of this CCC device,
- 1.0
mn: 2B.mm i.d.. 260-mLCapacity,Gear Side -..ant System: CHCI.:CH,COOH :0.1N HCI (2:2.., Stationary Phase: Nonaqueous Phase Sample: DNP Andm Acids Sample Volume: 10 mL DNP-Glu R m I U l m 400 rpm DNP-Asp Flow Rate 60 mUh
- 0.9 - 0.8 - 0.7 g g - 0.6
- 0.3 .- 0.2
DNP-beta4 DNP-Val
-
DNP-PI
- 0.1 10
5
- 1 0 0
15
10
Time (k
Time (h)
.
.,. ..,".
gure B. -...I amino acid separation with the combined hwizmtal' I coil (a) analytical oeparatlon: (b) peparatlve separation. Reprintedfrom Ana/. Biodxrm 1979, 100,271 542A
ANALYTICAL CHEMISTRY, VOL. 56, NO. 4. APRIL 1984
0.7
dfDNp-(C-\ hNP-delta10.6
planet centri.,,,
0.5
= 2
New Meffler PE balances have a formula for nrecise formula weighing; DeltaRange. Two new Mettler PE balances have the capacity to weigh big quantities and the readability to measure small ones. The PE360 and PE3600 with DeltaRangea,.
One PE wffh DeSiaRange does the work of two ordinary balances. Accurately measuring small amounts into large ones used to take two separate balances. One with the capacity to weigh the entire formula and another with the readability to precisely measure each ingredient. But DeltaRange lets you measure individual substances as you're
mixing the formula. A touch of the control bar increases accuracy tenfold, changing a readability of 0.01 g to 0.001 g. So even individual drops can be weighed as they're added to the mixture. Nothing could be easier. Or more accurate.
Transfers hformation to a computer. But DeltaRange is only one of the innovations these Mettler PE's offer you. Calibration is done automatically just by touching the control bar. Weighing values are always stable, vibrations being filtered out by a microprocessor.And PE CIRCLE 148 ON READER SERVICE CARD
Menler DeIlaRange brings a new degree of weighing aCCura~ytoyourlab
balances interface with computers, printers and other peripheral equipment for even greater flexibility in the laboratory. The Mettler PE360 and PE3600 begin a new generation of Mettler DeltaRange balances. They are priced at $1550 and $1475 respectively. For more information write to Mettler Instrument Corporation, Box 71. Hightstown,NJ 08520. Or call (609) 448-3000.
I 1 '
Flgure 10. Principle of flow-through centrifuge free of rotary seals (scheme V) Repinted tmm Reference 14
tershaft to the stationary pulley on the motor housing causes counterrotation of the countershaft with respect to the rotary frame. This motion is then conveyed to the central bowl by 1:l gear coupling, which doubles the angular velocity of the bowl. The motion of the countershaft is also transferred to the tube-supporting hollow shaft by means of 1:l pulley coupling. Thus. the svstem satisfies all reauiremen& indigated in Figure 10. Tl;e central bowl accommodates the toroidal coil around i v periphery (b)so that a
. _I_._ 11.
radially acting centrifugal force field establishes a hydrostatic equilibrium of the two-phase solvent system in the coil. This scheme produces stable retention of the Stationary phase of aqueous-aqueous polymer phase systems, which in other apparatus tend to form emulsions, and is particularly suitable for partition of cells and cell organelles (16.17). Nonsynchronous flow-through coil planet centrifuge (18).The mhst recent model of this versatile CCC centrifuge is shown in Figure 12. The apparatus has a rotary-seal-free design based on scheme VI in Figure 5. The main motor (left center) drives the rotary frame at a high revolutional speed, up to loo0 rpm, whereas the second motor (left lower) introduces slow rotation of the column holder at any desired rate between 0 and 50 rpm. Although the design of the apparatus becomes highly complex, a freely adjustable rotational rate of the coiled column permits unique applications which cannot be attained by other CCC schemes. The present model has been successfully applied to separations of macromolecules and cells using polymer phase systems and also to cell elutriation with single-phase physiological solutions. Unilateral HDES Recently, remarkable advances in
CCC technology have resulted from the discoverv and use of the unilateral hydrodynamic equilibrium of the two solvent phases in the rotating coil (19). The principle and mechanism of
this new CCC system are elucidated in Figure 13in which coiled columns are schematically drawn uncoiled to show the relative distribution of the two phases along the length of the column. In a rotating coil (top), the two solvent phases establish a hydrodynamic equilibrium in which the distribution ratio of the two phases is determined by the rotational speed of the coil. At a slow rotation (left), the two phases distribute evenly on the head side of the coil while any excess (not shown) of either phase remains at the tail. In this basic hydrodynamic system described earlier, the flow of either phase from the head to the tail of the coil at an optimum rate permits the retention of the other phase in nearly 5090 of the column space. However, further increase of the flow rate lowers the retention of the stationary phase, which eventually limits the application of a higher flow rate of the mobile phase. When the rotational speed of the coil is increased, the hydrodynamic equilibrium is gradually altered, and at the critical rotational speed (right) the two phases move toward opposite ends of the column and are completely separated, with one phase entirely occupying the head side and the other phase occupying the tail side of the column. For this mode of operation it is convenient to refer to these as the head phase and the tail phase. This unique hydrodynamic behavior of the two phases can be effectively used for performing CCC in three ways:
Design of flow-through centrifuge free of rotary seals (scheme V)
(a) aawr-ssdional view of th8 mtor; (b) peparaiar column vim taoidal coil arrangeman. Rsprlmed horn Reterence 15
544A
ANALYTICAL CHEMISTRY. VOL. 56, NO. 4, APRIL 1984
(continued on p . 548 A )
The Logical Suspect
c
The Logical Suspect Soot particle growth as it takes place in wood-burningfireplaces, diesel engines, and industrialfurnaces, has been attributed to a complex set of interdependent chemical reactions. A researcher at the General Motors Research Laboratories has demonstrated that the decomposition of a single species is primarily responsible.
P Soot Formation
S
OOT FORMATION may be
divided into two stages. Microscopic soot particles are generated in the “inception” stage. They reach full size in the “growth” stage, which accounts for more than 95%of their final mass. Most scientific exploration has concentrated on article inception which, despite a 1 the effort, remains unexplained. Dr.Stephen J. Harris, a hysical chemist at the General h?otors Research Laboratories, has reversed traditional priorities. Combining experiment with logic, he has formulated the first quantitative explanation of the growth stage in soot formation. Dr. Harris arrived at his mechanism through an elaborate process of elimination. To focus on the chemistry of soot growth, he began by eliminating from his
P
I
Ethylene/Oxygen (Mole R a t i d
Figure I : Total growth rate mntrosted wit) growth rate per unit mea plotted as a function of ethylene/oxy en mole mlio measured at a given height &e the bume: face.
Fzpre 2: AHisfS rendition of thesutfocegmwth of a single swt porticle by the incorporation of acetylene molecules.
investigation the complexities introduced by turbulence and mixing. He limited his research to premixed, ethylene/oxygen, laminar flames with one-dimensional flow. Previous descri tions in the literature told him t at two processes take place simultaneously during growth. Incipient particles collide and coalesce into larger particles, while growing at the Same time by incorporatin hydrocarbon molecules from the umed gases. The first process reduces total surface area without changin total mass, while the second, ca led “surface growth:’ increases both total surface area and total mass. Hence, the increase in the total mass of soot can be entirely attributed to surface growth. Dr. Harris set out to identify the hydrocarbon moleculesor “growth species”-responsible for surface growth. Increasing by increments the richness of the flame, he made the key discovery that although the total mass growth rate (gm/sec) increases strongly when the ratio of ethylene to oxy en is increased, the mass growt rate er unit surface area ycmz/secl’inaeases only sligh (see Figure 1) Thus, the contro&ng variable for how much soot is formed is not the concentration of growths ecies, but the surface area availa le for growth. This finding led him to con clude that richer flames roduce more total soot because tRey gen-
R
%
‘i
e;
\
II
TIME-
54SA
-
ANALYTICAL CHEMISTRY, VOL. 56, NO. 4, APRIL 1984
II
erate more g i c l e s in the inception stage. ore incipient particles offer greater inihal surface area for the incorporation of hydrocarbons. Since the growth rate per unit area must depend on growth species concentration, this concentration must be similar from flame to flame. Dr. Harris went on to reason that there must either be enough growth species at the outset to account for the total soot owth in the richest flame, or species must be rapidly formed within the flame from another hydrocarbon present in high enough concentration.
f&
H
E NARROWED his search
to the four most abundant classes of hydrocarbons found in flames: acetylene, polyacetylenes, 1 clic aromatic hydrocarbons , and methane. Methane can be eliminated, because its concentration does not decrease as soot is roduced. There is not enough P H to account for soot formation in an flame. Neither of these two hyC iocarbons can be readily formed from the other major species present. That left only acetylene and the polyacetylenes. Acetylene contains enou h hydrogen to a m u n t for the hy 0gen content of soot measured in the early stages of g r b t h . But among the polyacetylenes, only diacetylene could possibly supply enough h drogen. That left acetylene andrdiacetylene.
Fl8
1
&
There is more than enough acetylene to account for the mass of soot produced. There is not enough diacetylene and while diacetylene can be formed from the abundant suppl of acetylene, the reported rate o conversion is too slow for diacet lene to play a significant role. hat left only acetylene. Dr. Harris verified that acetylene is the growth s ecies by determining that the sliglt increase in growth rate per unit area is proportional to the increase in acet lene concentration (see Fig ure He also found that the rate constant he measured was in agreement with the reported rate constant for the decom osition of acetylene on carbon. hese findings confirmed his hypothesis that soot particles grow in flames by the incorporation and subsequent decom osition of acetylene. ‘560~that we know how soot grows:’ says Dr. Harris, “we can examine how it begins with greater understandin Then, perhaps our knowledge be complete enough to suggest better ways to reduce
Y
+
1,.
.F
d
soot:
General Motors
-
GM
THE MAN BEHIND THE WORK Dr. Stephen J. Harris is a Staff Research Chemist at the General Motors Research Laboratories. He is a member of the Physical ChemistrkDepartment. Dr a m s aduated from UCLA in 1971. received his Master’s and Ph.D. de ees in hysical chemistry from arvard bniversit y. His doctoral thesis concerned Van der Waals forces between molecules. Following his Ph.D. in 1975, a Miller Institute
f6
8
Motors in 1977. Dr. Harris conducted his investigation into soot particle growth with the aid of Senior Science Assistant Anita Weiner. His research interests at GM also include the use of laser diagnostic techniques in combustion analysis, with special emphasis on intracavity spectroscopy.
CIRCLE e4 ON READER SERVICE CARD
ANALYTICAL CHEMISTRY, VOL. 56. NO. 4. APRIL 1084
5471,
Figure 12. Recent model of the nonsynchronous flow-through coil planet centrifuge freeof rotary seals (scheme VI)
-
Repinled hwn RehXenCd 1
-
Basic
Hyd
Head
e
tmic Equilibrium
One Helical Turn
Flow
Tail
k
Head One
nic Equilibrium
Helical Turn
+
Tail
-
Normal Elution Mode
E s n Modi
Dual Countercurrent Syster
Sample Feed
548 A
ANALYTICAL CHEMISTRY, VOL. 56, NO. 4. APRIL 1984
Normal elution mode. The column is entirely filled with the head phase followed by flowing of the other phase from head to tail. Reversed elution mode. The column is entirely filled with the tail phase followed by flowing with the other phase from tail to head. Dual countercurrent mode. The column is filled with either phase followed by simultaneous elution of the two phases at the respective ends of the coil as illustrated. This requires an additional flow tube at each end of the coil to collect the effluent and, if desired, the sample feed port may be placed at the middle portion of the coil. This dual countercurrent system is open to future development.
Development of High-speed CCC Both the normal and reversed elution modes described above have the capability of producing a high retention of stationary phase far exceeding 50% of the total column space under conditions of moderate flow rate and have yielded relatively efficient separations (20). However, the operation of these schemes under unit gravitational field fails to maintain stable retention of the stationary phase against a high flow rate of the mobile phase. To sustain a high retention of the stationary phase, it is desirable to enhance the acceleration field by centrifugation. The search for a suitable system which produces a unilateral hydrodynamic equilibrium of the two phases under a strong centrifugal force field has been conducted by applying various types of planetary motion to various coiled columns (Figure 14). Scheme I synchronous planetary motion described earlier (Figure 5 ) was found to be inadequate because it gives rather even distribution of the two phases as in the basic hydrodynamic system. Various orientations of the coiled column were then examined using scheme IV synchronous planetary motion. Eccentric orientation on the holder of either parallel or toroidal form columns produced alternating segments of the two phases as in the basic hydrostatic system. Finally, when the tubing was coaxially wound around the holder, the ideal unilateral phase distribution was observed. Furthermore, the system was found to provide excellent retention of the stationary phase and efficient mixing of the two phases. With a short singlelayer helix the system has been successfully used for countercurrent extraction ( 1 9 , Z I ) . To increase both the resolving power and the sample-loading capacity of the system, a long column was accommodated compactly around a spool-shaped holder by wrapping the column in multiple
B&J Brand High Purity 1,I ,2-Trichlorotrifluoroethane makes normal phase HPLC convenient and accurate. Nonflammable and free of trace hydrocarbons, it is miscible with mobile modifiers such as methanol and acetonitrile. Get details on this and other B&J Brand High Purity Solvents by calling or writing today.
LABORATORIES,INC. 1953 South Harvev Street, Muskeaon. Michiaan U.S.A. 49442 616-726-3171 SubsidGary 01 American Hospital Supply Carporallon
CIRCLE 28 ON READER SERVICE CARD
B&J Brand High Purity Carbon Tetrachloride assures accurate trace hydrocarbon analysis by infrared spectroscopy. It’s quality controlled with a 25 mm path cell. For more details on how this and other B&J Brand High Purity Solvents can improve the accuracy of your laboratory procedures, write or call today.
I
LABORATORIES,INC.
1953 SouttI Harvey Street, Muskegon.Michigan U.S.A.49442 616-726-3171 Subsldlary 01 American timpttai SUPPIY corporamn
CIRCLE 30 ON READER SERVICE CARD
B&J Brand High Purity Water is purified and packaged by proprietary methods to assure consistent purity. Ideal for gradient elution HPLC, it eliminates “ghost” peaks and the variables of in-house produced water. Call or write for details on how B&J Brand High Purity Water and other B&J Brand high purity solvents can benefit your laboratory.
1-
1953 South Harvey Street Muskegon,Michigan U.S.A.49442 616-726-3171 Subsidiary 01 American Hospltal Supply C o r p a t i o n
CIRCLE 29 ON READER SERVICE CARD
B&J Brand High Purity Petroleum Ether meets the most demanding standards for pesticide residue analysis and gas chromatography. Consistent, reliable purity eliminates interfering background peaks, assures accurate results. Call or write for details on B&J Brand Petroleum Ether and other high punty solvents for trace analysis.
LABORATORIES,INC. 1953 South Harvey Street, Muskegon.Michigan U S k 49442 616-726-3171 Subsldtary 01 American tiospltal Supply Carporallon
CIRCLE 31 ON READER SERVICE CARD
ANALYTICAL CHEMISTRY, VOL. 56, NO. 4, APRIL 1984
549A
Figure 15. Prototype of the high-speed CCC apparatus with a multilayer coil layers. This multilayer coil has produced, as expected, highly efficient chromatographic separations even using a high mobile phase flow rate. A compact tabletop model (17 X 17 X 17 in.) of the high-speed CCC apparatus is shown in Figure 15. The motor (left, bottom) drives the rotary frame around the central stationary pipe by means of a pair of toothed pulleys and a toothed belt. A multilayer coil holder (lower position) and a counterweight (upper position) are symmetrically positioned on the rotary frame at a distance of 10 cm from the central axis of the apparatus. Coupling between the stationary sun gear on the central pipe and the identical planetary gear mounted on the holder shaft produces scheme IV synchronous planetary motion of the col-
by separation of pigments in the endclosed coiled tubes. Several years later, the two basic CCC systems were introduced; both are capable of continuous elution, monitoring, and fractionation as performed in liquid chromatography. The basic hydrostatic system was first improved to give rise to several efficient CCC schemes such as droplet CCC and locular CCC operating under a unit gravitational field and a helix CCC with the toroidal coil centrifuge employing a radial centrifugal field. A centrifugally operated droplet CCC apparatus has also been introduced recently (23).Although the basic hydrodynamic systems operating under a unit gravitational field produced efficient separations, the overall performance of the system was remarkably improved by application of a centrifugal force field. A variety of rotary-seal-free flow-through centrifuge schemes have been developed for subjecting the coiled columns to planetary motions. With synchronous planetary motion, scheme I produced efficient analytical separations, whereas scheme IV yielded efficient preparative separations. These two schemes were combined to form a versatile horizontal flow-through coil planet centrifuge. Several other useful CCC devices were also developed from the synchronous schemes including the angle rotor CCC (scheme II), the elution centrifuge (scheme 110,and the toroidal coil planet centrifuge (scheme IV). The most versatile nonsynchronous flow-through coil planet centrifuge (scheme VI) has been successfully constructed and applied to separations of macromolecules and cells with Summary polymer phase systems. The most recent breakthrough in The development of CCC technoloCCC technology came from the combigy over the past 20 years is summanation of scheme IV synchronous rized in Figure 17. The coil planet centrifuge first dem- planetary motion and the concentric coil orientation on the holder. This onstrated the potential of the method
umn holder. The revolutional speed of the apparatus is continuously adjustable up to loo0 rpm with a speed controller (left, above the motor). The initial multilayer coil was prepared from a single piece of PTFE tubing, 130 m long and 1.6-mm i.d. with a total capacity of about 285 mL. A pair of flow tubes from the multilayer coil (enclosed in a protective plastic tube) passes through the center hole of the holder shaft (right) and through a hole in the side of the rotating coupling pipe to enter the central stationary pipe. These flow tubes are not twisted by revolution as described in Figure 5 (scheme IV) and allow continuous elution through the rotating column without the use of rotary seals. The present model has been used for separations of various biological samples with excellent results (22).In Figure 16, chromatograms of DNP amino acids obtained by droplet countercurrent chromatography (DCCC) and high-speed CCC are compared. Both separations were performed with the standard solvent system composed of chloroform, acetic acid, and 0.1 N hydrochloric acid ( 2 2 1 ) and show similar peak resolutions. However, the elution of the DNP-proline peak required about three days using DCCC, whereas the elution of the same peak by high-speed CCC tmk place within 3 h. With the development of highspeed CCC, CCC should no longer be considered a slow separation method. It is expected that the separation times will be further shortened upon refinement of the high-speed CCC apparatus.
Figure 16. DNP amino acid Separations obtained by droplet CCC (a) and high-speed CCC (b) Part (a) reprinted horn Reterem 7: Part (b)repintedhan W w e m 22
550A
ANALYTICAL CHEMISTRY. VOL. 56, NO. 4, APRIL 1984
ROBOTICS = The Next Step in Vxboratory Automation
I
nas drastically shortened separation times without sacrificing peak resolution. The system has been successfully applied to countercurrent extraction with a short coil and to high-speed CCC with a multilayer coil. Dual countercurrent extraction and dual high-speed CCC are open to future development.
References (1) Ito, Y.J.Biochem. Biophys. Met. 1981,5,105. (2) Mandava, N. B.; Ito, Y.; Conway, W.D. Am. Lab. 1982,14,62. (3) Mandava, N. B.; Ito, Y.;Conway, W. D. Am. Lob. 1982,14,48. (4) Ito, Y.;Weinstein, M. A.;Aoki,L; Harada, R.; Kimura, E.; Nunagski, K. Noture 1966,212,985. (5) Ito, Y.; Aoki, 1.; Kimura, E. Anof. Chem. 1969,41,1579. (6)Sci. Ito,1970,8,315. Y.;Bowman,R. L. J. Chromtogr. (7) Tanimura,T.;Pisano, J. J.;Ito,Y.; Bowman, R. L. Science 1970.169,54. (8)Ita, Y. J. Chromatogr. 1980,188,33. (9) Ita, Y. J. Chromtogr. 1980,188,43. (10) Ito, Y.;Putterman, G. J. J. Chromatogr. 1980,193,37. (11) Ito, Y.;Bowman, R.L. Anol. Biorham _,._ .... 1975 fi6
Figure 17. Summary of the development of CCC technology over the past 20 years HSES. hyaostatic equilibrium system: M S . hYaDdyMmic equilibrium system; CCC. munter-nl chaatogaphy, DCCC. &oplet CCC: Hccc, helix CCC; LCCC. Imlat CCC; RLCCC. m i m LCCC; UCCC, gyration LCCC; cpc,m i l planet m i f u s e ; rs.. m i n g sea
210
(12) Ito, Y.;Bowman,R.L.; Noble, F. W. A w l . Biochem. 1972,49,1. (13) 1to.Y. J. Chromtoar. 1980.192.75. (14) Ito, Y.; Suaudeau, 57; Bowman,R. L. Science 1975,189,999. (15) Ita, y,; L.Anal, B,o. ehem. 1978,85,614. (continued on p. 554 A )
Here’s why IBM HPLC Columns give you superior reproducibility and resolution
IBM Packing: contded size and uniformly shaped silica spheres provide maximum adsorption or partition surface. Flow patterns are uniform for high efficiency and outstanding reproducibility.
Ordinary Spherical Packing: non-uniform size and imperfect shape result in variable surface and flow characteristics. Both resolution and reproducibility suffer.
IBM gives you premium quality HPLC Columns without a premium price. To order or for further information, call 800-213-3122. In Connecticut call collect, 265-5791. Ask for Dept. 34. Or write IBM Instruments, Inc., Dept. 34,PO Box 3020, Wallingford, CX 06492 CIRCLE 110 ON READER SERVICE CARD
552A
ANALYTICAL CEMISTRY. VOL. 56. NO. 4, APRIL 1984
Irregular Shaped Silica Packing: adsorbent surface and flow patterns can vary widely. Fractured particles can collect and block flow. Column life is shortened.
A powerful microprocessor makes the QMG 064 the most versatile, easy to use RGA available. Compare these benefits: 0 Full mass spectnun (1-64 amu) to fingerprint your vacuum system using the scan mode- the ultimate diagnostic tool. 0Monitor eighr gasses sequentially with the cycle mode
AH *is for the Pfiw of an ion Wge! The QMG W is one of a family of quadrupole mass spectrometers manufactured by Balzers. All i n s t r u t s are backed by a worldwide service organization. If you have a requirement for vacuum instrumentation, call or write:
feature.
0Measure tom1 pressure or pawl pressures of eight user sekaed gasses with the touch of a butron. M
A
AA
J
116) Sutherland. 1. A,; Heywood-Wadding.
ton. D.; Pe1ers.T. -1. J. Liy. Ckromatocr. 1984.7.:im. 117) Vianagan, S. I).; dohanrron, G.; Yost, H.; Ito; Y.; Sutherland, 1. A. J. Liy. (!hmmatogr. 1984,7,:IH5. (18)lto, Y.; Hramhlett, C . T.; Rhatnagar,
Bio~Radnow offers carbo-
H,O
hydrate chemists a choice of eight complementary HPLC columnsto provide better analysis of dairy products, juices. fermentation broths, wood pulp hydrolysatesand othercarbohydrate mixtures. Each of these eight columns is dedicated to a specific analysis: I Aminex Fast Carbohydrate (100mm.)-Rapid monosaccharides I Aminex'"'HPX-87P- sugars, cellulose hydrolysates I Aminex HPX-87C (250mm.)-sugar alcohols I AminexHPX-S7C(300mm.) -monosaccharides I Aminex HPX-42C-Rapid oligosaccharides
~" " ' . ~ \ vnur'
'
Equally important, Bio-Rad providesin-depthapplications assistance. If you have a problem with carbohydrateseparations, or wish to optimize them, giveour Applications Staff a call. Or contact us at:
. .
Crw'
E':: Bia Rad Laboratories
I
I
b
Aminex-HPX-87HCarbohydrates in solution with acids, fermentation monitoring I Eio-Sil" Amino-5S -Disaccharides
I
R.; Huherrnsn, M.; Leive, I.. l,.;Culiinane. I.. M.;(;roves, W. SPP.Sci. Tech. 19113,18, :
