Get documents about "Fused Core Particles for HPLC Columns
Document Sample
Fused Core Particles for HPLC
Columns
by Joseph J. Kirkland, Timothy J. Langlois,
and Joseph J. DeStefano
In recent years, there has been a
strong movement toward the use of
HPLC columns with smaller particles.
While columns with 5-µm particles
have long been the standard, some
users now have turned to columns of
3.5 µm1 or smaller2 to produce faster
separations and increase sample
throughput. The advantages of smaller
particles are well known: faster separa-
tions without sacrificing resolution so
that more samples can be analyzed in
the same time period. Of course, there
are always tradeoffs that must be made
to gain this speed. Smaller particles
decrease the permeability of columns;
thus increased pressures must be used,
resulting in increased stress on pump- Figure 2 Particle size distribution of 2.7-µm fused core particles.
ing systems and other instrument
components. In addition, columns
with particles of ^3 µm must usually filtering samples and mobile phases is tion and the higher density of the fused
be fitted with end frits of 0.5-µm or often a chore that small-particle users core particles. The end result is a col-
1.0-µm porosity to ensure that smaller must endure. umn that produces very fast separations
particles in the particle distribution for high sample throughput with con-
do not escape to cause detection and This article describes a 2.7-µm fused ventional equipment and user-friendly
column life problems. These smaller- core particle (Advanced Materials operating techniques. The thin outer
porosity frits are subject to fouling by Technology, Inc., Wilmington, DE) shell of the particles permits very rapid
microparticulates from samples and that has been designed to allow very fast solute mass transfer (optimized kinetics)
mobile phases, eventually causing col- separations without some of the disad- so that high mobile phase velocities can
umn failure because of very high back- vantages of conventional columns with be used for fast separations without a sig-
pressures. To eliminate this problem, small, totally porous particles. The char- nificant loss in column efficiency.
acteristics of these fused core particles
represent a fortunate compromise of
separation speed with modest operating Materials and methods
pressures. In addition, however, because The fused core particles were synthe-
of the unique structure of a thin porous sized in the authors’ laboratory using
shell surrounding a uniform solid silica proprietary nanoparticle technology.3
core, columns of these materials exhibit The highly purified Type B silica parti-
unusually high efficiency. While tradi- cles are 2.7 µm in total diameter with a
tional commercial columns of totally 0.5-µm-thick outer shell of 90-Å pores,
porous particles rarely show reduced resulting in a nitrogen surface area of
plate heights, h (plate height/particle G150 m2/g. The fused core particles
diameter) of less than 2, columns of the have been given the trademarked name
fused core particles often exhibit h values of Halo™, which suggests their physical
of about 1.5 for molecules with molecu- configuration with the spherical porous
lar weights up to at least 300. This shell covering a solid core (Halo fused
unusually high efficiency is believed core columns available from Mac-Mod
Figure 1 Electron micrograph cross- by the authors to be a function of the Analytical, Chadds Ford, PA). A cross-
section of 2.7-µm fused core particle. extremely narrow particle size distribu- section electron micrograph of one of
18 APRIL 2007 • AMERICAN LABORATORY
the particles is shown in Figure 1. Fused Results and discussion
core Halo particles are synthesized with
an extremely narrow particle size distri- The unusually high efficiency of
bution, as indicated in Figure 2. fused core Halo columns is illus-
trated by the data in Figure 3. Here,
Densely bonded monofunctional the reduced plate height minimum
dimethyl-C8 and dimethyl-C18 station- for the small molecule, naphthalene
ary phases were prepared on the fused (MW 128.2), is about 1.5, represent-
core particles by conventional reactions4 ing about 12,000 theoretical plates
with silanes obtained from Gelest, Inc. for this column. Very little increase
(Morrisville, PA). These bonded phase in plate height occurs for this small
materials were subsequently endcapped molecule when the mobile phase
with trimethylsilane groups. Carbon/ velocity is more than doubled. This
hydrogen analyses were by Micro- effect is the result of the favorable
Analysis, Inc. (Wilmington, DE). The Figure 3 Comparison of reduced plate height versus kinetics of the thin porous shell on
level of bonded stationary phase was sim- mobile phase velocity plots for naphthalene and loraz- the fused core Halo particles. The
ilar to that for totally porous particles and epam. Column: 50 × 4.6 mm, Halo C8, 2.7 µm. Sol- time for solute diffusion in and out
in keeping with that expected for densely utes: naphthalene (MW = 128.2), mobile phase: 60% of the thin porous shell containing
bonded materials, where additional silane acetonitrile/40% water; lorazepam (MW = 321.2), the stationary phase is minimized
groups cannot be further reacted because mobile phase: 30% acetonitrile/70% 20 mM sodium so that band broadening is also
of steric limitations. For example, the phosphate buffer, pH 3.5. Temperature: 24 °C. minimized as the mobile phase
bonded phase level on one sample of is increased. Note also in Figure 3
Halo C18 was 3.29 µeq/m2 prior to end- that the reduced plate height for the
capping. Particle size measurements were were performed using 1-µL injections drug lorazepam (MW 321.2) at the plate
conducted on a Gemini V instrument from a model 8125 sampling valve height minimum is also about h ~1.5 for
(Micromeritics, Norcross, GA) and elec- ( Rheodyne, Cotati, CA). The bed column efficiency equivalent to that for
tron microphotographs were obtained stability study was performed with the smaller molecule, naphthalene. As
from Micron, Inc. (Wilmington, DE). the model 1100 instrument using a the mobile phase velocity is increased
Surface areas were conducted with a model 1100 series automatic sampler for lorazepam, the plate height increases
Multisizer 3 Coulter Counter (Beckman (Agilent Technologies). The 5-µm slowly as mobile phase velocity increases
Coulter, Fullerton, CA). particle Ace column was from Mac- as a result of the slower diffusion of the
Mod Analytical, the 3.5-µm particle much larger molecule. Yet, at the highest
Chromatographic data were Zorbax XDB-C18 column was from mobile phase velocity (almost 8 mm/sec
obtained on a model 1100 instru- Agilent Technologies (Wilming- at 328 bar), lorazepam shows a reduced
ment ( Agilent Technologies, Palo ton, DE), and the 2.5-µm XBridge plate height of about 2.5, indicating
Alto, CA) or on a model SPD-6A C18 column was from Waters Corp. that the 50-mm column is still operat-
liquid chromatograph (Shimadzu (Milford, MA). Plate heights were ing with about 7000 theoretical plates.
Scientific Instruments, Tokyo, calculated with the data systems It is believed that the unusually high effi-
Japan) using UV detection with a using the peak half-height/width ciency of fused core Halo columns is the
2-µL cell. Data were recorded from method.6 Data for the reduced plate result of the extremely narrow particle
the model 1100 ChemStation (Agi- height versus mobile phase velocity size distribution and the 30–50% higher
lent) or from an in-house-designed plots in this paper were fitted to the density for the fused core particles (rela-
computer system using a model 900 Knox equation: tive to totally porous particles). These
interface (PE Nelson, Cupertino, properties apparently allow the forma-
CA). All data were acquired with h = Au1/3 + B/u + Cu (1) tion of a more homogeneous packed bed,
a detector response time of 0.1 sec, which results in efficiencies higher than
in addition to a data sampling rate where h is the reduced plate height for conventional totally porous packings
of at least 20 points/sec so that (plate height h/particle size dp); A, B, (reduced A term, Eq. [1]). While reduced
a minimum of 20 points would be and C are column coefficients; and u plate heights of 2 for columns is generally
obtained on the very sharp, low-vol- is mobile phase velocity.6 accepted as a practical limit, the possi-
ume peaks obtained with the fused bility of increased efficiency for packed
core Halo columns. Stainless steel Solvents for mobile phases were from HPLC columns as a result of optimized
column hardware and column frits Mallinckrodt Baker (Phillipsburg, NJ). particles was predicted by Knox.7 In
with 2-µm porosity were from Isola- Test solutes were from Sigma-Aldrich the Knox study, reduced plate heights
tion Technologies, Inc. (Hopedale, (St. Louis, MO). Lorazepam and van- of less than 2 (some as low as 0.6) were
MA). The 4.6 × 50 mm and 2.1 × tin were obtained by dissolving single obtained with large, solid glass beads and
50 mm columns used in this study commercial pills in the appropriate superficially porous particles. The higher
were prepared by slurry packing mobile phase. No effect from inactive density and narrow particle size distribu-
techniques using in-house-designed excipients in these pills was observed tion of the fused core particles appear to
hardware. 5 Chromatographic tests in the authors’ studies. be the characteristics that allow reduced
AMERICAN LABORATORY • APRIL 2007 19
FUSED CORE continued
plate heights of less than 2 to
be a common feature.
Predictably, as shown in Fig-
ure 4, the smaller (2.7 µm) par-
ticle size of the Halo particles
results in significantly higher
efficiency for lorazepam than
5-µm and 3.5-µm particle col-
umns. The less steep increase in
plate height with mobile phase
velocity increase for the Halo
particles is also in keeping with Figure 7 Separation of test mixture. Column:
Figure 4 Effect of particle size. Columns: 50 × 4.6 the smaller particle size and the 4.6 × 50 mm, 2.7-µm Halo C8. Mobile phase: 75%
mm, Ace 5-C18, 5 µm; Zorbax XDB-C18, 3.5 µm; superior kinetic properties of methanol/25% 20 mM potassium phosphate buffer, pH
Halo C18, 2.7 µm. Solute: lorazepam (MW = 321.2). the fused core structure. The 7.0. Flow rate: 1.5 mL/min. Temperature: 25 °C.
Mobile phase: 30% acetonitrile/70% 20 mM sodium higher efficiency of the smaller- Solutes: 1) uracil, 2) butyl paraben, 3) propranolol, 4)
phosphate buffer, pH 3.5. Temperature: 24 °C. particle fused core column does naphthalene, 5) acenaphthene, 6) amitripylene.
come with the price of a higher
operating pressure. However,
the operating pressure at the
highest mobile phase velocity
used for a very fast separation is
still within the pressure limits
of most of the HPLC instru-
ments currently in use.
Relative to results for a col-
umn of 2.5-µm hybrid parti-
cles, the 2.7-µm particle Halo
column shows significantly
better performance for the
large drug molecule, vantin
Figure 5 Comparative column performance for Figure 8 Separation of aromatic acids. Column:
(MW 557.6), as illustrated in
vantin. Columns: 50 × 4.6 mm, 2.5-µm XBridge 4.6 × 50 mm, 2.7-µm Halo C8. Mobile phase: 55%
Figure 5. Not only is the plate
C18 (k = 9.5), 2.7-µm Halo C8 (k = 8.7). Solute: methanol/45% 25 mM sodium phosphate buffer, pH
height about 67% smaller, but
vantin (MW = 557.6). Mobile phase: 35% acetoni- 2.5. Flow rate: 2.20 mL/min. Temperature: 24 °C.
the increase in plate height
trile/65% 20 mM sodium phosphate buffer, pH 3.5. Column pressure: 360 bar. Solutes: 1) uracil, 2)
with mobile phase veloc-
Temperature: 24 °C. phthalic acid, 3) 2-fluorobenzoic acid, 4) 3-nitroben-
ity increase is less, due to
zoic acid, 5) 3-fluorobenzoic acid, 6) m-toluic acid.
enhanced kinetic properties
of the fused core structure.
The mechanical strength of the prepared prior to the final sample
fused core particles, coupled with injection to ensure a mobile phase
a narrow particle size distribution that was equivalent to the starting
and higher particle density, also mobile phase. After these 500 sample
allows unusually stable column injections at a flow rate of 1.0 mL/min
beds to be formed by the slurry that resulted in 260 bar, the column
packing method. Figure 6 shows showed no change in solute retention
the results of a column stability or efficiency, within the precision of
study for a 2.1 × 50 mm fused making up the mobile phase.
Figure 6 Halo column stability study. Column: core Halo C8 column. Here, the
2.1 × 50 mm, 2.7-µm Halo C8. Mobile phase: 50% column was operated for 71 hr Examples of separations with fused
acetonitrile/50% water (premixed). Flow rate: 1.0 with continuous premixed mobile core Halo columns are shown in
mL/min. Column pressure: 260 bar. Temperature: phase flow (over 40,000 column Figures 7 and 8. The high column effi-
24 °C. Solutes: 1) uracil, 2) phenol, 3) 4-chloro-1- volumes) with solvent recycle. ciencies and good peak shapes are the
nitrobenzene, 4) naphthalene. Solid line: first chro- During this period, 500 sample result of the favorable particle con-
matogram; dotted line: chromatogram after 71 hr of injections were made on the col- figuration, densely bonded surfaces,
continuous flow (>40,000 column volumes) and 500 umn with an automatic sample and the very high purity of the Type B
sample injections. injector. Fresh mobile phase was silica that make up the particles.
20 APRIL 2007 • AMERICAN LABORATORY
Conclusion 4. Snyder, L.R.; Kirkland, J.J. Introduc- 7. Knox, J.H. J. Chromatogr. A 1999,
tion to Modern Liquid Chromatography, 831, 3.
Fused core particles for HPLC columns 2nd ed., John Wiley and Sons: New
have been developed that are well York, NY, 1979; Chapter 5.
suited for very fast separations and high The authors are with Advanced Materials
5. Kirkland, J.J.; DeStefano, J.J. J. Chro-
sample throughput at modest column matogr. A 2006, 1126, 50. Technology, Inc., 3521 Silverside Rd.,
backpressures. The 2.7-µm high-purity 6. Snyder, L.R.; Kirkland, J.J.; Glajch, Ste. 1-K, Quillen Bldg., Wilmington, DE
silica particles have a solid core and J.L. Practical HPLC Method Develop- 19810, U.S.A.; tel.: 302-477-2513;
a 0.5-µm-thick outer shell with 90-Å ment, 2nd ed., John Wiley and Sons: fax: 302-477-2514; e-mail: jkirkland@
pores, providing a surface area of 150 New York, NY, 1997; Chapter 2. advanced-materials-tech.com.
m2/g. The unusually high efficiency for
columns of these particles is believed to
be a feature of the very narrow particle
size distribution and the higher particle
density. Reduced plate heights, h, of
~1.5 for small molecules represent a
level of efficiency that has previously
not been reported for stable, commer-
cial HPLC columns. The thin outer
porous shell on these fused core par-
ticles allows rapid solute mass transfer
(fast kinetics) so that column efficiency
ACTIVA-M
is degraded very little as mobile phase
velocity (flow rate) is increased. These
particles can be slurry packed into very A NEW
stable column beds, and continuous
use of pressures up to at least 400 bar DIMENSION
is indicated since the column bed has
been carefully stabilized. Samples and
in
mobile phases can be treated in the
same manner as for columns of 5-µm ICP Analysis
particles since 2-µm frits are used on HORIBA Jobin Yvon presents the Master
the inlet and outlet, enhancing column of CCD-based ICPs for true multi-line analysis
lifetime compared to columns of ^3-
µm particles.
As for most systems, operation at higher
temperatures (e.g., 40–60 °C) will further
increase column efficiency and decrease - New proprietary
column backpressure. For best results, the ICP-based database
high-efficiency columns of the fused core - Unique interactive
particles should be used in high-quality
instruments with low-volume fittings, assistance tools
injectors, and low-volume detectors to - Full benefit of multi-line
minimize extra-column band broaden- analysis
ing. Data handling systems should be
used that capture at least 20 data points/
sec (using detector response times of 0.1 Bring ICP expertise
sec) to maintain peak integrity for fast,
low-volume (sharp) peaks. to your lab with an
easy-to-use instrument.
References Be confident in the quality of
1. Kirkland, K.M.; McCombs, D.A.;
your results.
Kirkland, J.J. J. Chromatogr. A 1994,
660, 327.
2. Kirkland, J.J. J. Chromatogr. Sci.
2000, 38, 535.
3. Kirkland, J.J.; Langlois, T.J. Substrates www.jobinyvon.com
with Porous Surfaces. Patent applied
for, Feb. 2006.
Web access AL20.com?3706
AMERICAN LABORATORY • APRIL 2007 21
Related docs
