Volume VI
Issue 1, 1996
Strategy For Faster Methods Development Using
Symmetry Columns and pH Optimisation
®
Zoubair El Fallah
The selectivity prism shown here demonstrates the method development strategy for
ionisable analytes. pH is the primary tool for the development of the separation, solvent
selectivity is used to fine-tune the separation.
Introduction
As ICH (International Conference on
Harmonisation) and FDA guidelines
for impurities specification in drug
substances and products are likely to
become more stringent, methods
development in the pharmaceutical
laboratory is becoming an even more
challenging task. Demands to increase
productivity in today’s analytical labo-
ratories, are coupled with requirements
to enhance the sensitivity and the
ruggedness of assays. Progress in both
HPLC instrumentation and column
chemistries has given analytical meth-
ods the speed, specificity, accuracy
assessment to a sophisticated software
program. In addition, one also has to
invest time in understanding the theory,
assumptions and limitations that are
built into the software, along with the
methodology used to assess separation
quality and experimental design. It is
true that experimental design using a
software package is more likely to be
a rational one, however, knowledge-
able chromatographers may require
fewer experiments to reach the same
optimum as that of a computer-assisted
Table of Contents
Faster Methods Development.................1-5
Use of Narrow-Bore Columns............. 6-15
Triazine Herbicides and Metabolites ...16-19
Formaldehyde in Workplace Air ........ 20-23
Literature Corner............................. 24-25
An Interview
and sensitivity that have never before
with the inventors of EnantioSelect ......... 26
been achieved, making HPLC the tech-
Whats new ........................................ 27
nique of choice for pharmaceutical
assays. However, method development How to reach us ................................. 28
and validation is still a laborious and
Application Notes
costly process. It is to a large extent still
Paraquat/Diquat
conducted on the basis of trial and
error, although methods development Carbonyl Compounds as DNPH
Derivatives, EPA Method 554
software packages are readily avail-
EPA Method 604, Phenols by HPLC
able in the market. The resistance that
these software packages have encoun-
Editor: Uwe Neue
tered is most likely due to the fact that
ISSN # 1084-0540
chromatographers prefer to rely on their
visual judgment and experience rather
than delegate the process of optimisa-
tion and chromatographic quality
1
optimisation (1). An intermediate
approach to methods development
would take advantage of both the chro-
matographer’s intuition and experience
and the theory of chromatographic
retention. This is the essence of the
method development strategy presented
here. It is based on the use of
Symmetry ® columns (2) with pH and
solvent composition optimisation.
Since the majority of pharmaceutical
compounds are ionisable, selectivity is
more likely to be affected by changes
in the pH of the mobile phase.
Therefore, the possibility of working
over the entire pH range without
adversely affecting the performance of
the system is an important consideration
in method development. This is espe-
cially true when the solutes are of basic
character. With conventional reversed-
phase HPLC packings, the addition of
modifiers to block the silanols is neces-
sary to achieve good peak shape,
especially when operating at neutral
pH. This has an impact on both the
sensitivity and the ruggedness of the
method, which demonstrates that
column quality and performance often
dictate the strategy for method devel-
opment. With Symmetry columns, one
can use simpler mobile phases with
low-to-moderate buffer concentrations,
and still achieve good peak shapes,
even with strong bases. This makes it
possible to devise a simple strategy for
methods development that is primarily
based on pH. Often, pH turns out to
be the principal parameter affecting
the separation. However, analysts tend
to examine its effect only when solvent
optimisation fails. We show that consid-
ering pH first, significantly speeds
up the process of method development.
Theory
Gradient elution chromatography is
known to be a powerful tool for the
analysis of complex mixtures. It is also
the method of choice when dealing
with partially or totally unknown sam-
ples, which is the case when impurity
profiling is performed on a drug sub-
stance and/or product. Gradient chro-
matography makes it possible to
achieve a higher peak capacity per unit
time (3), and it has a better sensitivity
than isocratic mode, as band concen-
tration often occurs. However, in the
pharmaceutical laboratories,
gradient methods are avoided due to
the fact that they are very sensitive to
the type of instrumentation used, and
therefore, they are not easy to transfer
and to establish as routine rugged ana-
lytical methods. Isocratic methods are
preferred but they suffer from the fact
that for relatively complex samples, the
sensitivity of the assay is relatively low,
especially for late eluting peaks. This
effectively limits peak capacity.
In reversed-phase HPLC, retention times
obtained from linear solvent strength
gradients are related to isocratic elution
data and to parameters that depend on
the nature of the solute and solvent in
use. This relationship between gradient
and isocratic retention is based on a
Linear Solvent Strength theory intro-
duced by Snyder (4,5). Using this
theory, it is possible to predict retention
times under isocratic or other gradient
conditions. The theory is based on the
assumption that the variation of the ln k
is linear with the volume fraction of the
organic modifier:
ln k = ln k 0 – S (Eq.1)
where S is the slope of that variation
and k 0 is the retention factor in pure
water.
The retention time of a given peak
under gradient conditions is related
to S and k0 through the following
relationship:
tg = t 0 + td + (1/S.b).
ln(1+ S.b.t 0.k 0.exp(-S. Ø0)) (Eq.2)
t 0 is the retention time of an unretain-
ed solute, td is the delay time of the
HPLC system, b is the slope of the gra-
dient (% organic change/min) and Ø0
is the organic composition at the begin-
ning of the gradient.
In order to describe the retention of a
given compound in the mixture, one
needs to know its corresponding para-
meters, k 0 and S, in Eq.1. Since it is
not practical to perform isocratic runs
to derive these parameters, due to the
fact that one needs to guess what com-
position would lead to a reasonable
retention window, the use of gradient
retention data, as shown in Eq.2, is a
better alternative. Two gradient runs
using the selected solvent are needed
in order to estimate k 0 and S, for each
one of the peaks in the chromatograms.
2
These gradient runs differ in their
slopes; that is, b1 and b2 while keeping
all other conditions the same. The first
gradient conditions (i.e., Ø0 and b1,
have to be selected so that all peaks
are eluting primarily in the gradient por-
tion; this supposes a relatively low Ø0,
and that enough resolution is obtained
between all peak pairs or peaks of
interest, which means that b1 should be
around 0.5 to 1% organic change per
minute. b2, is typically either twice or
half b1, depending on the retention
range of the first run.
The combination of two sets of Eq.2;
that is, two different gradient runs,
leads to the following relationship
which can be numerically solved for S:
b2 exp(tr 1.S.b1 – 1) -
b1 exp (tr 2.S.b2 – 1) = 0 (Eq.3)
tr i is the reduced gradient retention time,
obtained with the slope bi:
tr i = tgi – t 0 – td (Eq.4)
Once S is solved, ko can be obtained
from the following relationship:
k 0 = (1/ to.S.b1)(exp (tr 1.S.b1 +SØ0 )
- exp(SØ0 )) (Eq.5)
Therefore, for a given peak, the varia-
tion of k vs. fraction of organic is
obtained from retention data of two
gradient runs by solving Eq.3 for S, and
obtaining k0 from Eq.5.
This methodology of performing two
runs is used by software packages like
DryLab™ (LC-Resources)
HIPAC™ (Phase Separations)
Experimental Section
The Waters HPLC system consisted
of the following: A 616 LC System,
a 717plus Autosampler, a 996
Photodiode Array Detector. Data acqui-
sition was done by the Millennium®
2010 Chromatography Manager.
For the separation of the mixture of
antidepressants and their metabolites,
a Symmetry C18 5µm (3.9 mm I.D. x
150 mm) column was used. Column
temperature was maintained at 35 °C.
Eluent A was either 20 mM phosphate
buffer at pH 2.3 or pH 7.0 (MilliQ
water, Millipore), while solvent B was
methanol. The flow rate was 1 mL/min.
The effluent was monitored at 254 nm.
Results and Discussion: Figure 1: Effect of PH on the separation of 2 antidepressants and their metabolites.

We present a simple example that Peak Identification:

Gradient:
demonstrates the power of pH as a Start: 20% MeOH 1: Nordoxepin
primary tool of optimisation. The sam- Slope: 1%/min. 2: Doxepin
ple consisted of amitriptyline and dox- Flow rate: 1mL/min. 3: Nortriptyline
4: Amitriptyline
epin, and their respective metabolites,
nortriptyline and nordoxepin. These a - 20 mM potassium phosphate buffer at pH 7.0
0.25 4
solutes are known to be strongly basic,
and they tend to exhibit poor peak 0.20
1 3
shape, if eluted at all, on conventional 0.15 2

AU
reversed-phase columns. 0.10
0.05
Since Symmetry columns give good
peak shapes independent of pH, 0.00

we are able to explore the selectivity - 0.05

0 5 10 15 20 25 30 35 40 45 50 55 60
changes due to pH first, before per-
forming any solvent composition optimi- b - 20 mM potassium phosphate buffer at pH 2.3
0.15
sation. Figure 1 shows a comparison of 4
2 1 3
two chromatograms obtained at pH 0.10
2.3 and pH 7.0, both with 20 mM
AU

phosphate buffer. It is clear that there is 0.05

a strong change in selectivity between
the two gradient runs. At pH 7.0 the 0.00
components are more strongly retained, 0 5 10 15 20 25 30 35 40 45 50 55 60
which is expected with basic solutes, Minutes
and they are also evenly spaced. At
pH 2.3 however, we have less reten-
tion and the peaks are bunched in Gradient Chromatograms obtained at neutral and acidic pH are compared for overall
pairs, and with a wide space between quality of the separation.
the pairs. In addition, an elution order
reversal occurred. Figure 2: Estimation of solvent composition for 10 minutes retention time.
Now we have to make some judgment
on the quality of the chromatography
of these two runs. It is clear that the 10
selection of pH 7.0 will more likely Nordoxepin Nortriptyline
lead to a better separation than pH 8 Doxepin Amitriptyline
2.3. Therefore, our next gradient run
6
was performed at pH 7.0, but now
71% MeOH
having a gradient slope of 2%/min. 4
The combination of the retention times
of these two gradient runs will allow us 10 min Retention Time
2
to derive the relationship between
retention factor and mobile phase com- 0

position, for each one of these com-

-2
pounds, as explained above. Table I 0.00 0.20 0.40 0.60 0.80
summarises S and k0 obtained based
Fraction of Methanol
on the two gradient runs at pH 7.0.
Figure 2 shows the variation of ln k
vs. methanol composition for the four Variation of In(k) vs. MeOH Volume fraction. Linear model based on data from two
solutes. It is interesting to note that the gradient runs at pH 7.0.
ternary amines, doxepin and amitripty-
line, have similar S values, while the Table 1
secondary amines, nordoxepin and
nortriptyline also have similar values. Peak Nordoxepin Doxepin Nortriptyline Amitriptyline
A goal for this assay was an analysis
k0 230.4 1224.2 1326.1 6185.7
time of under 15 minutes.
S 8.29 9.40 8.44 9.63
When we set up our criteria for optimi-
sation based on a total run time of

3
10 minutes (i.e., retention time of
amitriptyline, the last eluting solute),
we reach the conclusion that 71%
MeOH should be selected. However,
at this point we have to examine two
important things:
– do we have enough retention for
the first eluting peak?
– do we have enough resolution
between all peak pairs?
Based on S and k0 values for nordox-
epin, we found that its retention factor,
k at 71% MeOH, is 0.64. Selectivity
values between adjacent peak pairs
were also calculated at 71% MeOH,
based on values reported in Table I,
and were found to be 2.4, 2.1 and
2.0, respectively. With today’s
columns, these selectivities are expect-
ed to give baseline separation. Figure 3
shows the separation obtained at 71%
MeOH/29% 20 mM phosphate buffer
at pH 7.0. As expected, separation of
the four solutes was achieved with a run
time of about 11 minutes. However, the
retention factor of nordoxepin is much
higher than predicted, and selectivity
values were 2.0, 1.3 and 2.1, respec-
tively. There are several reasons that
could account for the difference
between predicted and observed
selectivities:
– the linear model is only an approxi-
mation, and quadratic behavior has
been observed (6). This is especially
true with ionisable solutes, as they may
elute under a mixed mode retention.
– errors in estimating delay volume will
affect early eluting peaks more than
late eluters.
– there are inherent proportioning
errors, typically 0.5%, both during gra-
dient and final isocratic composition.
Exact prediction is not the purpose of
the strategy presented here. Data from
gradient runs are used only to guide us
to the retention window.
Figure 3: Separation of two antidepressants and their metabilites at pH 7.0.
71% Methanol / 29% 20 mM potassium phosphate buffer.
0.25
1 Nordoxepin
1 2 Doxepin
0.20
3 Nortriptyline
2 4 Amitriptyline
0.15
3
AU
4
0.10
0.05
0.00
0 2 4 6 8 10 12 14
Minutes
The final chromatogram is obtained using the predicted mobile phase of 71% methanol
and 29% 20 mM phosphate buffer at pH 7.
Conclusion:
Often it is necessary to include pH impurity profiles and assays for degra-
as a criterion in method development. dates. We have shown that following
Gradient runs at two different pH this strategy, the use of Symmetry
values (acidic and neutral) in the early columns can significantly reduce the
stage of the optimisation are very time of methods development.
useful in directing the analyst towards
the right pH.
A second gradient run is then per- References
formed at the selected pH, and its data
(1) A. Tchapla, P. Allard and S. Heron, LC-GC Intl. 6
can be successfully used to locate the (1993), 86.
optimum composition, based on time
(2) U.D. Neue, D.J. Phillips, T.H. Walter, M. Capparella, B.
or/and resolution criteria. Once the pH Alden and R.P. Fisk, LC-GC 12 (1994), 468.
and retention range are selected,
(3) J.M. Davis, J.C. Giddings, Anal. Chem. 55 (1983),
mobile phase mixture optimisation can 418.
be carried out, when necessary. In
(4) L.R. Snyder, In “ High Performance Liquid
order to carry out this strategy success- Chromatography, Advances and Perspectives”, Vol.1, C.
fully, a column with good performance Horvath, ed., Academic Press, New York, (1980), pp 207-
at acidic as well as neutral pH is nec- 316.
essary. This was a simple example (5) L.R. Snyder, J.W. Dolan and J.R. Grant, J. Chromatogr.
demonstrating the basic approach. 165 (1979), 3.
We have applied this strategy to signifi- (6) P.J. Schoenmakers, H.A.H. Billiet, R. Tijssen and L. De
Galan, J. Chromatogr. 149 (1978), 519.
cantly more complex samples, such as
For your free copy of the
Symmetry Application Notebook,
please check box 3 on the Business
Reply Card and return today.
4
Ordering Information
Symmetry Analytical Steel Columns
Chemistry Particle Size Dimensions Part No. fi Automated gradient methods optimisa-
Symmetry C8 5 µm 2.1 mm x 150 mm WAT056955
tion assumes excellent integrity in HPLC
Symmetry C8 5 µm 3.0 mm x 150 mm WAT054230 system gradient generation. Efficient,
Symmetry C8 5 µm 3.9 mm x 150 mm WAT046970 precise and accurate gradient formation
Symmetry C8 5 µm 4.6 mm x 150 mm WAT045995 is a key criteria for HPLC solvent delivery
Symmetry C8 5 µm 4.6 mm x 250 mm WAT054270 systems, one in which Waters instruments
Symmetry C18 5 µm 2.1 mm x 150 mm WAT056975 excel. In Waters 616 LC System, unique
Symmetry C18 5 µm 3.0 mm x 150 mm WAT054200 gradient control software provides reliable,
Symmetry C18 5 µm 3.9 mm x 150 mm WAT046980 reproducible results.
Symmetry C18 5 µm 4.6 mm x 150 mm WAT045905 Waters patented Random Phase
Symmetry C18 5 µm 4.6 mm x 250 mm WAT054275 Synchronisation (RPS™) Software monitors
both eluent composition and flow rate,
Symmetry Steel Cartridge Columns (All cartridge columns require reusable endfittings.) adjusting automatically and instantaneously
Chemistry Particle Size Dimensions Part No. to current operating conditions to ensure
Symmetry C8 5 µm 3.9 mm x 50 mm WAT054240 optimal performance. Combined with a low
Symmetry C8 5 µm 3.9 mm x 150 mm WAT054235 system volume of <400 µL, changes in the
Symmetry C8 5 µm 4.6 mm x 150 mm WAT054255 eluent blend are quickly and accurately
Symmetry C8 5 µm 4.6 mm x 250 mm WAT054245 delivered to the HPLC column to facilitate
Symmetry C18 5 µm 3.9 mm x 50 mm WAT054220 fast separations for rapid methods
Symmetry C18 5 µm 3.9 mm x 150 mm WAT054205 development efforts.
Symmetry C18 5 µm 4.6 mm x 150 mm WAT054210
Symmetry C18 5 µm 4.6 mm x 250 mm WAT054215

Symmetry Validation Kit (three steel columns or steel cartridges from 3 different batches of packing materials)
Chemistry Particle Size Dimensions Part No. For more information on
Symmetry C8 column 5 µm 3.0mm x 150 mm WAT054434 Symmetry columns, please
Symmetry C8 column 5 µm 3.9 mm x 150 mm WAT046955 check box 4 on the Business
Symmetry C8 column 5 µm 4.6 mm x 150 mm WAT054435
Reply Card and return today.
Symmetry C8 column 5 µm 4.6 mm x 250 mm WAT054438
Symmetry C18 column 5 µm 3.0 mm x 150 mm WAT054446
Symmetry C18 column 5 µm 3.9 mm x 150 mm WAT047210
Symmetry C18 column 5 µm 4.6 mm x 150 mm WAT054448
Symmetry C18 column 5 µm 4.6 mm x 250 mm WAT054450 Symmetry Columns are also available
Symmetry C8 cartridge 5 µm 3.9 mm x 150 mm WAT054440 in narrow-bore. See article on page 6.
Symmetry C8 cartridge 5 µm 4.6 mm x 150 mm WAT054442
Symmetry C8 cartridge 5 µm 4.6 mm x 250 mm WAT054444
Symmetry C18 cartridge 5 µm 3.9 mm x 150 mm WAT054452
Symmetry C18 cartridge 5 µm 4.6 mm x 150 mm WAT054454
Symmetry C18 cartridge 5 µm 4.6 mm x 250 mm WAT054456

Sentry™ Guard Columns* and Guard Holders

Particle Size Quantity Dimensions Part No.

Symmetry C8 5 µm 2/pkg 3.9 mm x 20 mm WAT054250

Symmetry C18 5 µm 2/pkg 3.9 mm x 20 mm WAT054225
Integrated Guard Holder
(for Waters steel cartridge columns** only) WAT046905
Universal Guard Holder (for any HPLC column) WAT046910

Part No.
End Connector Kit
(includes 2 sets of reusable endfittings, 2 C-Clips and Kalrez o-rings) WAT037525

* Sentry guard columns can be used with either the steel cartridge columns but require the appropriate Symmetry, Sentry, RPS, and Waters are trademarks of
guard holder. Waters Corp.
** For Waters 3.9 mm x 50 mm cartridge column, you must use the Universal Guard Holder. DryLab is a registered trademark of LC Resources, Inc.
HIPAC is a registered trademark of Phase Separation, Ltd.

5
Performance of Symmetry Narrow-Bore
Columns for Isocratic and Gradient Analyses
Dorothy J. Phillips
The Symmetry ® columns are offered in
the traditional analytical column diame-
ters (4.6 mm and 3.9 mm) as well as
in narrow-bore columns (3 mm and
2. 1 mm). The four column diameters
give the flexibility needed for method
development and validation. Choosing
a column geometry involves a compro-
mise. As the column diameter is varied,
there are trade-offs between resolution
and sensitivity, resolution and capacity,
speed and back pressure. The smaller
diameter columns offer increased sensi-
tivity, shorter analysis times, and
reduced solvent consumption. Gradient
methods can be transferred to narrow-
bore columns, if the system delay vol-
ume is considered.
Introduction
Narrow-bore columns are now being
used to reduce solvent consumption, to
increase sensitivity, and to meet the
need for analysis of small sample vol-
umes. An increasing need for narrow-
bore columns is to interface liquid chro-
matography with other analytical tech-
niques (such as LC/MS). Symmetry
packings are offered in 2.1 mm and
3 mm I.D. columns to meet the request
for reduced solvent consumption and
analysis of small sample volumes. The
3 mm I.D. column offers the benefits of
narrow-bore columns without the need
to make significant instrument changes.
In contrast, the 2.1 mm column requires
modification to the HPLC system for
optimum performance.(1)
Narrow-bore columns with the same
packing as a standard analytical col-
umn should give the same selectivity
and resolution when all columns are
operated at equal linear velocity. (2)
The resolution is affected by the peak
width which is determined by band dis-
persion in the stationary phase. The
peak widths in time units should be the
same in all columns; the peak volumes,
however, will be proportional to the
back pressure
solvent
speed consumption sensitivity
Scheme 1
cross-sectional area, and hence to the
flow rate. A review of the fundamental
principles of peak dispersion confirms
the expectation for equal efficiencies of
the peaks regardless of column diame-
ter. The three major causes of band dis-
persion in a packed bed are covered
by the van Deemter equation:
H = A + B/u + Cu where H is the
theoretical plate height in the column
and u is the linear velocity.
A is related to multipath dispersion,
B to longitudinal diffusion and C to
resistance to mass transfer of a solute.
Neither term of the equation is depen-
dent on column diameter. Therefore,
the band width due to dispersion in
the packed bed and the resulting reso-
lution should be independent of
column diameter.
Yet increased band dispersion has been
noted for the 2.1 mm column on tradi-
tional HPLC instruments, which has limit-
ed the use of small diameter columns
(1, 2). A decrease in efficiency has
been noted when the diameter is
reduced. The additional dispersion is
due to dispersion in the tubing and to
the volume from the injector to the
detector which is termed extra-column
volume. The smaller the peak volume,
the larger is the impact of the extra-col-
umn volume. The total width of a peak
is a combination of the width due to the
band in the column and the width con-
tributed from the HPLC system
(dispersion and volume).
6
resolution flow rate
injection column system
volume diameter volume
Scheme 2 Scheme 3: Gradient
w 2total = w 2 + w2
column system
where w is the peak width
For the traditional analytical column of
4.6 mm I.D., it is acceptable for the
system to contribute 100 µL to the peak
volume, since a typical peak may be
as large as 300 µL or more. However,
a sample separated on the 2.1 mm col-
umn can have components with total
peak volumes less than 100 µL, there-
fore, the extra-column volume must be
minimised. Extra-column volumes of
25 µL to 45 µL can be achieved by
using minimum length tubing for con-
nections and/or smaller diameter tub-
ing (changing from 0.009 in to 0.005
in I.D.) and by putting a microbore cell
in the detector. In summary, the differ-
ences in peak efficiencies between the
different diameter columns are due
mainly to extra-column volume.
This report also covers the advantages
of the narrow-bore columns as well as
the compromises that have to be made
during their use. Data are presented for
the new Symmetry C18 narrow-bore
columns and the traditional analytical
columns. The column length was not
varied in this study; all columns are
150 mm long. The differences in assay
speed and solvent consumption for the
different column diameters as well as
the relative back pressures will be
addressed (Scheme 1). The perfor-
mance of the four different diameters
will be discussed in terms of
compromises to meet resolution and
sensitivity needs (Scheme 2). Transfer
of gradient methods from traditional
analytical to narrow-bore columns
will be discussed in terms of system
demands and corrections for system
volume (Scheme 3). This report also
discusses the use of the Waters
Sentry™ Guard column with the
narrow-bore columns.
Experimental
The HPLC system consisted of the 625
LC System, a 717plus Autosampler,
and a 486 Tunable UV/Vis
Absorbance Detector. The Millennium®
2010 Chromatography Manager was
used for system control and data acqui-
sition. System suitability software was
used for data analysis. The convention-
al HPLC system band spread (extra-
column band spread from injector to
detector cell) at 0.25 mL/min was
90 µL. The microbore system had extra-
column band spread of 40 µL with the
microbore cell in the detector (2.7 µL
volume). The column is removed from
the system to determine the extra-col-
umn band spread; acetone is used as
the marker. The band spread is deter-
mined by measuring the peak width, in
minutes, for acetone at 4.4% of peak
height and multiplying this value by the
flow rate to convert the units to volume.
The samples and chromatographic
Table 1: Mobile Phase and Sample Preparation
Analysis
Plate Count
Barbiturates
Sulfa Drugs
Alberta Peptides
Tamoxifen
conditions are listed in Table 1. Flow Results and Discussion
rates for all analyses were 0.6 mL/min
The four different diameter columns
for 3 mm I.D. columns and 0.29
gave good efficiencies for a neutral
mL/min for the 2.1 mm column. The
marker. The 4.6 mm and 3.9 mm I.D.
3 mm Symmetry C18 columns were also
columns gave 15,000 and 13,500
used at 1 mL/min and compared to
plates for acenaphthene, respectively.
the 3.9 mm and 4.6 mm columns. All
The 2.1 mm I.D. Symmetry C18 columns
four diameter Symmetry columns
had average efficiency for acenaph-
(2.1 mm, 3 mm, 3.9 mm, and 4.6 mm)
thene of 10,700 plates. The
were compared at equal linear velocity
3 mm Symmetry C18 column efficiencies
(flow rates based on square of column
for acenaphthene averaged 11,420
diameter).
plates.
Alberta Peptide Mixture
Scheme 1: Speed, Back Pressure and
The peptide mixture is from Alberta Solvent Consumption
Peptide Institute at the University of
Typical back pressure and solvent
Alberta in Edmonton, Alberta, Canada.
requirements for the four column geome-
The sample is called the Peptide
tries are given in Figures 1 and 2 and
Performance Standard for reversed-
Table 2. Table 2 contains the data for
phase chromatography (Cat. RPS-
the sulfa drugs separated with a
P0010). The five decapeptides are
methanol/pH 2.9 acetic acid mobile
identified below. (3) S1 has three posi-
phase. The back pressures of these
tive (3+) charges and S2 to S5 have
columns, when operated at equal linear
two positive charges (2+). The sample
velocity with the 20% methanol mobile
is prepared by dissolving the lyophilised
phase, were 1960 psi and 2150 psi
mixture in 1 mL of 0.1% TFA or solvent
for 2.1 mm and 3 mm columns, respec-
A; this should give 100-200 injections
tively. The dramatic decrease in solvent
at 0.1 AUFS at 210 nm.
consumption observed with the narrow-
bore columns is immediately apparent
Ac-Arg-Gly-X-X-Gly-Leu-Gly-Leu-Gly-Lys-Amide
S1 Arg-Gly-Ala-Gly- Gly-Leu-Gly-Leu-Gly-Lys-Amide
S2 Ac-Arg-Gly-Gly-Gly-Gly-Leu-Gly-Leu-Gly-Lys-Amide
S3 Ac-Arg-Gly-Ala-Gly-Gly-Leu-Gly-Leu-Gly-Lys-Amide
S4 Ac-Arg-Gly-Val-Gly-Gly-Leu-Gly-Leu-Gly-Lys-Amide
S5 Ac-Arg-Gly-Val-Val-Gly-Leu-Gly-Leu-Gly-Lys-Amide
Mobile Phase Sample Preparation
acetonitrile/water acenaphthene, 0.6 mg/mL and
60:40 (v/v) acetone, 6 µL/mL
100 mM potassium phosphate pH 7/ eight component mixture
acetonitrile/water 20:30:50 (v/v/v) with 25 µg/mL each
water/methanol/glacial acetic acid six component mixture with
79:20:1 (v/v/v) 10-39 µg/mL each
Solvent A: 0.1% TFA in water five component peptide mixture
Solvent B: 0.1% TFA, in acetonitrile
Gradient: 10% to 40 %B in 30 min.
50 mM potassium phosphate pH 3/ 600 µg/mL of mobile phase
acetonitrile 20:30:50 (v/v)
7
from Table 2. The 3 mm column
required 56% less solvent than the
4.6 mm column for analysis of the sulfa
drugs. The 2.1 mm column requires
80% less solvent per analysis than the
4.6 mm column.
Figure 1 shows the chromatography of
the sulfa drugs on the different
Symmetry columns at equal flow rate.
At 1 mL/min flow rate, the smaller the
column diameter, the higher the back
pressure; the back pressure increases
from 1900 psi for the 4.6 mm column
to 3150 for the 3 mm column. The nar-
row-bore columns have the advantage
of shorter run times; the analysis is com-
plete in 5 minutes on the 3 mm column
and requires 10 minutes on the 4.6 mm
column. The analysis time could be
even shorter on the 3 mm column;
H 2N SO2NH2
however, the high back pressure limits 2000 psi for each. While the efficien-
the flow rate. The 2.1 mm column has cies show some decrease with column
an even more limited flow rate range diameter, the differences are small
due to high back pressures. The faster enough to be of no practical conse-
flow rates on the smaller diameter quence. The results obtained confirm
columns also result in a loss of efficien- that selectivity and resolution are inde-
cy since the columns are not being pendent of column diameter when all
operated at optimum linear velocity columns are operated at equal linear
(at minimum in van Deemter curve). velocity. Additional reductions in extra-
When the columns are operated at column bandspreading should make
equal linear velocity as shown in Figure the efficiencies for peak 5 even closer
2, the same chromatography can be in value.
obtained on the 2.1 mm, 3.0 mm,
3.9 mm and 4.6 mm I.D. columns. The
back pressures for the four columns are
similar in this case at approximately
Figure 1: Performance of Symmetry C18 5µm Columns: Equal Flow Rates
Column: a) Symmetry C18 4.6 mm x 150 mm
b) Symmetry C18 3.9 mm x 150 mm
c) Symmetry C18 3.0 mm x 150 mm
Mobile Phase: water/methanol/glacial acetic acid 79:20:1
3 Flow Rate: 1mL/min
1 4 Detector: 254 nm
5
Sample: mixture of 6 sulfa drugs
2 6

1: Sulfanilamide

H N a)
H 2N SO2N

N
0.0 2.0 4.0 6.0 8.0 10.0
2: Sulfadiazine
3
4 5 USP Plates USP Tailing Back Pressures
S
1
H2N SO2NH
Peak 4
N 2 6 a) 4.6 mm ID 7560 1.1 1900 psi
b) 3.9 mm ID 5710 1.1
3: Sulfathiazole c) 3.0 mm ID 3160 1.2 3150 psi

NH2 SO2NH
b)
N
CH3
4: Sulfamerazine 0.0 2.0 4.0 6.0 8.0

N
CH3 3 4 5
NH2 SO2NH
1 6
N 2
CH3
5: Sulfamethazine

S NHSO2 NHCOCH2CH2COOH c)
• H2O

N
0.0 2.0 4.0 6.0 8.0
6: Succinylsulfathiazole
Minutes

The smaller the column diameter, the higher the back pressure but the shorter the run time.
Efficiency decreases with column diameter.

8
Table 2: Performance of Symmetry® C18 5 µm Columns: Methanol as Organic Modifier

Column Column Flow Rate Analysis Mobile Phase Column

Dimensions Volumes Time per Analysis Back pressures

2.1 mm x 150 mm* 0.52 mL 0.29 mL/min 7.4 min. 2.1 mL 1960 psi (125 atm)

3.0 mm x 150 mm 1.06 mL 0.6 mL/min 7.4 min. 4.4 mL 2150 psi (137 atm)
1.0 mL/min 4.7 min. 4.7 mL 3150 psi (200 atm)

3.9 mm x 150 mm 1.79 mL 1.0 mL/min 7.4 min. 7.4 mL 2450 psi (156 atm)

4.6 mm x 150 mm 2.49 mL 1.0 mL/min 10.3 min. 10.3 mL 1900 psi (121 atm)
1.4 mL/min 7.6 min. 10.6 mL 2020 psi (129 atm)

Mobile Phase: water/methanol/glacial acetic acid 79:20:1

Sample: mixture of six sulfa drugs
* The 2.1 mm column was operated on a system with a microbore flow cell in the detector, 40 µL extra-column band spread.

Figure 2: Performance of Symmetry C18 5µm Columns: Equal Linear Velocities

3
4 5
1 Column: a) Symmetry C18 4.6 mm x150 mm
6 b) Symmetry C18 3.9 mm x150 mm
2
A c) Symmetry C18 3.0 mm x150 mm
d) Symmetry C18 2.1 mm x150 mm
Mobile Phase: water/methanol/glacial acetic acid 79:20:1
Flow Rates: a) 1.4 mL/min
b) 1.0 mL/min
c) 0.6 mL/min
0.00 10.00 d) 0.29 mL/min
3 Sample: mixture of 6 sulfa drugs, 10 to 39 µg/mL
4 5 Injection volumes: a) 14 µL
1 b) 10 µL
2 6
c) 6 µL
d) 3 µL

b)

0.00 10.00

3
4 5
1 6
2 USP Plates USP Tailing
Peak 4
a) 4.6 mm ID 5780 1.1
b) 3.9 mm ID 5710 1.1
c) c) 3.0 mm ID 4350 1.2
d) 2.1 mm ID 4140 1.2

0.00 10.00

3 5
4

1 6
2

d)

0.00 10.00
Minutes

At equal linear velocities, the run times are the same and the efficiencies are similar. The narrow-bore columns offer
reduced solvent consumption and require less sample to achieve equivalent detector response.

9
Sentry™ Guard Column Table 3: Effect of Sentry Guard Column on 3 mm I.D. column
The Sentry Guard columns (3.9 mm x
Compounds USP Plates USP Tailing Back Pressures
20 mm) were developed for use with
Factor
the 3.9 mm and 4.6 mm I.D. columns.
The fact that the guard column has the Acenaphthene
same or smaller I.D. than the analytical without guard column 10,650 1.2 1210 psi
column weighs heavily on its ability to with guard column 11,000 1.2 1300 psi
maintain or improve performance when
in line with the analytical column.(4) Butabarbital (peak 4 Fig. 3)
Therefore, we investigated the use of without guard column 7080 1.2 1610 psi
the Sentry Guard column with narrow- with guard column 7660 1.4 1760 psi
bore columns. As shown in Figures 3
and Table 3 the Symmetry C18 Sentry
Guard column did enhance the perfor-
mance of the 3 mm I.D. column with Figure 3: Effect of Sentry Guard Column on 3 mm I.D. Column
the same chemistry. The retention times
are 25% longer with the guard column Columns: a. Symmetry C18 3mm x 150mm
b. column a with Symmetry C18
as expected; the 20 mm addition to the
Sentry Guard Column (3.9mm x 20mm)
total column length led to a 23% Mobile Phase: 0.1M potassium phosphate, pH7/
increase in total column volume acetonitrile/water 20:30:50 v/v/v
(change 1.06 mL to 1.3 mL). The plate Flow Rate: 0.6 mL.min
Detection: 214 nm
counts for both acenaphthene in the
Injection Volume: 6µL of sample with each barbiturate
plate count test and for the barbiturates at 25 µg/mL
increased 3% to 8%. Therefore, the
Sentry Guard column can be used with a. Symmetry C18 3 mm column
the 3 mm I.D. columns. 3
1
2
0.20 4
H 5
O 7
O N O
C2H 3 6
HN
C2H3 CH2=CHCH2 0.10 8
NH
N O (CH3)2CHCH2
O
H O

1: Barbital 5: Butalbital 0.00

0.00 5.00 10.00
H H
O N O O N O Minutes
CH2=CHCH2 CH3CH2
NH NH b. 3 mm column with Sentry Guard Column
CH2=CHCH2 (CH3)2CHCH2CH2
O O 1 3

2: Allobarbital 6: Amobarbital 0.30 2

4
0.20 5
H CH3 7
O N O O N O 6
C2H3 CH3CH2 NH 0.10 8
NH

O O

0.00

3: Phenobarbital 7: Mephobarbital 0.00 5.00 10.00

Minutes
H H
O N O O N O The Sentry Guard Column causes an increase in analysis time but does not cause a
CH2=CHCH2
CH3CH2
NH NH decrease in efficiency; an improvement in resolution is noted.
CH3CH2CH CH3(CH2)2CH
O O
CH3 CH3

4: Butabarbital 8: Secobarbital

For more information on

Sentry Guard Columns, please
check box 5 on the Business
Reply Card and return today.

10
Scheme 2: Sensitivity, Load and Resolution Figure 4: Sensitivity as a Function of Column Diameter
Different column geometries are needed
to meet the user's separation demands 4
in terms of resolution, speed, load, sen-
3.5
sitivity and solvent consumption. (1) The
maximum concentration (Cmax) of sam- 3
2.1 mm x 150 mm
ple in the detector at the peak apex is 2.5
inversely proportional to the square of 2
the column diameter and the column
1.5
length (Cmax ¬ 1/(d2 x L)). Smaller 3.9 mm x 150 mm
diameter columns provide an increase 1

in sensitivity, if injection volume and 0.5

detector cell volume are kept constant. 0
Figure 4 plots relative sensitivity versus 0 1 2 3 4 5
flow rate and shows that the F (mL/min)
2.1 mm I.D. column gives fourfold
greater sensitivity for equivalent injec-
A decrease in column diameter from 3.9 mm to 2.1 mm gives a fourfold increase in
tion volume than the 3.9 mm I.D. relative sensitivity.
column. The peak volume is less on
the smaller diameter column, and thus
gives a higher concentration at peak Figure 5: Sensitivity versus Resolution with Different Column Diameters
maximum. The curve for the 2.1 mm
column ends abruptly at approximately Columns: Symmetry C18 5µm 150 mm length
1 mL/min because higher flow rates Mobile Phase: 50mM potassium phosphate.
pH 3.0/acetonitrile 55:45
result in excessive back pressures.
Flow Rates: a. 1.4 mL/min
Choosing a column geometry may b. 1.0 mL/min
involve a compromise. The need for c. 0.60 mL/min
d. 0.29 mL/min
high sensitivity has to be balanced Detection: 240nm
with the resolution demands. The injec- Sample: tamoxifen, 600 µg/mL, 7µL injection
tion volume can be kept constant while
the column diameter is reduced for 0.0015
increased sensitivity. However, if injec-
AU a. 4.6 mm I.D.
tion volume is too high for a given col- 12 3
umn volume, then a loss in resolution
0.0000
can result. The change in resolution
5.00 10.00
with increased sensitivity can be Minutes
unacceptable for the given method. 0.0015
Figure 5 shows the chromatography
for the impurity peaks in tamoxifen; AU 12 3 b. 3.9 mm I.D.

impurities at the 0.1% level or higher of

the tamoxifen peak detector response 0.0000
must be quantitated for regulatory com- 5.00 10.00
Minutes
pliance. The increase in peak height
demonstrates the gain in sensitivity that
0.0015
is obtained by using a smaller column
diameter at constant injection volume. 12 3 c. 3.0 mm I.D.
AU
However, in this assay, the resolution
between the peaks decreased with 0.0000
column diameter and became unac- 5.00 10.00
Minutes
ceptable on the 2.1 mm column; the
3
impurities could not be quantitated. 0.0015 2
1

AU d. 2.1 mm I.D.

0.0000
5.00 10.00
Minutes

At constant injection volume, the peak height increases as column diameter decreases.
The 2.1 mm column has the highest sensitivity but shows a loss in resolution.

11
Larger diameter columns have more
capacity, allowing separation to be
maintained at larger sample loads.
However, the larger diameter columns
require more sample than the narrow-
bore columns for a given sensitivity or
detector response. For the tamoxifen
assay, the 2.1 mm I.D. column requires
only 20 µL of a given concentration to
get the same signal as observed for
a 100 µL injection on the 4.6 mm I.D.
column.
For your copy of the Waters
Column Selection Guide, please
check box 6 on the Business Reply
Card and return today.
(CH3)2NCH2CH3O
C=C
Tamoxifen
Scheme 3: Gradient Analyses
The above discussion is straightforward
for isocratic analyses: one simply needs
to change the flow rates when working
with different diameter columns.
However, when gradient methods
are used, other factors should be con-
sidered. Here, the HPLC system deter-
mines if the transfer from a 4.6 mm to
the 2.1 mm column is seamless when
the columns are operated at equal lin-
ear velocity. If the HPLC system has a
significant delay volume (volume from
static mixer to column head >1 mL),
then the gradient will reach the 2.1 mm
column noticeably later than the larger
diameter column. Therefore, the onset
of the gradient will be delayed and
retention times will be significantly
longer on the narrow-bore column than
on the traditional column (4.6 mm I.D.
and 3.9 mm I.D.). Figure 6 shows this
effect on a system with a large delay
volume. The first peak is eluted at
about 9 minutes on the 4.6 mm column
and at 18 minutes on the 2.1 mm I.D.
column.
The elution times for the narrower-bore
columns can be brought closer to those
obtained using the analytical columns
by modification of the HPLC system.
The system delay volume needs to be
reduced. The Model 717plus
Autosampler can be modified by
installing the low delay volume kit and
using smaller diameter tubing through-
out the system. Removing the static
mixer for flow rates <0.5 mL/min and
using 0.009 in I.D. tubing from pump
to injector reduces the delay volume in
the Model 625 LC System. Figure 7
12
CH2COOH
• HO – C – COOH
CH2COOH
C2H5
shows how the gradient starts earlier
for the 2.1 mm I.D. column on the
modified 625 system than on the con-
ventional system; the differences in
retention times are now significantly less
between the 4.6 mm and 2.1 mm
columns (9 versus 12 minutes for peak
1, respectively). The newer Model 616
Solvent Delivery System has an even
lower delay volume (<800 µL) and
causes even less shifts in retention times
between the 4.6 mm and 2.1 mm
columns. The Model 616 can be oper-
ated with static mixers for the 4.6 mm
and 3.9 mm columns and without them
for both the 3 mm and 2.1 mm columns
(Figure 8). Removing the static mixers
does not affect the performance of the
Model 616 system up to 1 mL/min
flow rate.
Conclusions
The four column diameters give the
user the flexibility required to obtain
optimum loading, back pressure, sensi-
tivity, resolution and solvent consump-
tion. Symmetry narrow-bore columns
have the advantages of increasing
sensitivity, reducing solvent consumption
or decreasing analysis time. The 3 mm
I.D. column has an advantage over the
2.1 mm column in that the HPLC system
does not have to be modified for opti-
mum performance. Gradient methods
can be transferred from traditional
analytical columns to the narrow-bore
columns by using either systems with
low dwell volume or by reducing the
delay volume of the solvent delivery
system.
Figure 6: Effect of Column Diameter on Gradient Analyses at Equal Linear Velocity

– 0.15
a. 4.6mm I.D.

– 0.25

– 0.35
Columns: Symmetry C18 5µm columns; 4.6 mm, 3.9 mm,
10.00 15.00 20.00 25.00 3 mm and 2.1 mm I.D. x 150mm
b. 3.9mm I.D. Flow Rates: a. 1.4 mL/min,
b. 1 mL/min,
0.10
c. 0.59 mL/min,
0.05 d. 0.29 mL/min
Mobile Phase: A. 0.1% TFA aqueous
0.00 B. acetonitrile with 0.1% TFA
10.00 15.00 20.00 25.00 Gradient: 10% to 40% B in 30 minutes, step to 60%
B for 5 min, 100% A for 40 min
c. 3mm I.D. Sample: Alberta Peptide mix
0.10
a. 7µL
b. 5µL
0.05 c. 3µL
d. 2µL
0.00
Detector: 214nm
10.00 15.00 20.00 25.00 Temperature: 35 ˚C

d. 2.1mm I.D.

0.10

0.05

0.00
10.00 15.00 20.00 25.00
Minutes

The system delay volume causes the retention times to become longer as the column diameter decreases; at slower flow rates
it takes longer for Solvent B to reach the head of the column.

Figure 7: Effect of System Delay Volume on Gradient Analysis with Different Diameter Columns

a. 4.6mm I.D. conventional HPLC system

-0.15
5
Columns: Symmetry C18 5µm 4.6 mm x 150 mm
4
Symmetry C18 5µm 2.1 mm x 150 mm
1 3
-0.25 2 Flow Rates: a. 1.4 mL/min
b. 0.29 mL/min
c. 0.29 mL/min
-0.35 Mobile Phase: A. 0.1% TFA aqueous;
10.00 15.00 20.00 25.00 B. acetonitrile with 0.1% TFA
Minutes Gradient: 10% to 40% B in 30 minutes, step to 60%
5
B for 5 min, 100% A for 40 min
b. 2.1mm I.D. conventional HPLC system 4 Sample: Alberta Peptide mix
0.10 a. 7µL
3 b. 2µL
1 2
0.05 c. 2µL
Detector: 214nm
0.00 Temperature: 35 ˚C
Modified HPLC System:
10.00 15.00 20.00 25.00 Model 626 LC System without static mixer
Minutes and 0.009 in I.D./ tubing
c. 2.1mm I.D. modified HPLC System
4 5

0.10 3
1
2

0.05

0.00

10.00 15.00 20.00 25.00

Minutes

After modification of the HPLC System, shorter retention times were obtained with the 2.1 mm column compared to before the
modification. They were also more similar to what was obtained with a 4.6 mm column.

13
Figure 8: Gradient Analyses with Different Diameter Columns using the Waters 616 LC Delivery System

a) 4.6mm I.D.
0.10 Columns: Symmetry C18 5 µm columns;
4.6 mm, 3.9 mm.
3.0 mm and 2.1mm I.D. x 150 mm
AU

0.05 Flow Rates: a. 1.4 mL/min,

b. 1 mL/min.
c. 0.59 mL/min,
0.00
d. 0.29 mL/min
Mobile Phase: A. 0.1% TFA aqueous;
10.00 B. acetonitrile with 0.1% TFA
Minutes Gradient: 10% to 40% B in 30 minutes,
step to 60% B for 5 min,
b) 3.9mm I.D. 100% A for 40 min
0.10 Sample: Alberta Peptide mix
a. 7µL
b. 5µL
AU

0.05 c. 3µL
d. 2µL
Detector: 214nm
0.00
Temperature: 35 ˚C

10.00
Minutes

c) 3.0mm I.D. without static mixer

0.10
AU

0.05

0.00

10.00
Minutes

d) 2.1mm I.D. without static mixer

0.10
AU

0.05

0.00

10.00
Minutes

The low system volume of the Model 616 system gives similar chromatography regardless of the column diameter. The removal of the
static mixers further reduces the system delay volume.

References
1. U. Neue, "Selecting the Best Column Dimension for a Chromatography", LC Magazine 4(3) (1986), pp. 250-
Separation", Waters Column VI(1) 1993, pp. 1-5. 254.

2. "Small Bore Liquid Chromatography Columns: Their
Properties and Uses", R. P. W. Scott, editor, John Wiley &
Sons, New York, 1984.
3. C. T. Mant and R. S. Hodges, "Requirement for Peptide
Standards in Reversed-Phase High Performance Liquid
4. M. Capparella, W. Foster, M. Larrousse, D. Phillips, A.
Pomfret and Y. Tuvim, "Characteristics and Applications of a
New High Performance Liquid Chromatography Guard
Column", J. Chromatogr. A 691 (1995), pp. 141-150

14
Ordering Information
Symmetry Analytical Steel Columns

Chemistry Particle Size Dimensions Part No.

Symmetry C8 5 µm 2.1 mm x 150 mm WAT056955

Symmetry C8 5 µm 3.0 mm x 150 mm WAT054230

For more information on
Symmetry C18 5 µm 2.1 mm x 150 mm WAT056975

Symmetry C18 5 µm 3.0 mm x 150 mm WAT054200

Symmetry columns, please
check box 4 on the Business
Reply Card and return today.

Symmetry Validation Kit (3 steel columns from 3 different batches of packing materials)

Chemistry Particle Size Dimensions Part No.

Symmetry C8 column 5 µm 3.0 mm x 150 mm WAT054434

Symmetry C18 column 5 µm 3.0 mm x 150 mm WAT054446

fi Waters offers multiple system configurations to support narrow-bore HPLC separation

methods. The Discovery System is based on Waters 616 LC System, 717plus Autosampler,
and 996 Photodiode Array Detector. The system offers narrow-bore and analytical HPLC
methods capabilities for a full range of micro and standard analyses. Waters BioDiscovery
System, based on the 626 LC System, provides a non-metallic fluid path in a microanalysis
system. Ideal for biomolecule separations, this system enables use of optically clear mobile
phases that extend low-UV wavelength detection sensitivity. Both systems may operate
with high sensitivity, single-channel UV/visible detection capability with Waters 486 Tunable
Absorbance Detector. Incorporation of Waters photodiode array detection, offering spectral
and detection sensitivity at the picomole level, enables collection of spectral information
useful in determining component homogeneity and identity.

Rapid Po
lyolefin
Additive
Separat
Objective: ion Using
The object Nova-Pa
demon ive of
strate the this application k ®C
HPLC for utility of note
reversed-pha is to 18
Analysis protected large scale purific
synthetic a-tions se

AccQ • of™ Hy
oligonucleot of DMT 1.5 AU

ls: dro Detai ides.

Tag Me lyzed Bovin Synthetic

e Serum
oligonucleot
thod for used in
a
ranging wide variety
ides are
success-fully
Alb
10
Objective:
Amino of applic

Acid An umin Using

from use
utilization as hybrid a-tions 8
nm

and the as primers for ization Column

The object :
DNA sequeprobes to
e 230

Bondap
demon ive of
strate the this applicMost ationrecentalysis Waters
polymerase
chain reactio ncing
6
Eluent
A:
Radial-P
ak® HC18
ak™ cartridgHA (8 x 100mm
rbanc

HPLC for utility of effecti note ly,

veness
studies
is to have n* (PCR). 5
100mM
ammon e)
reversed-pha
protected large scale purific
HPLC Pu phosphodiesseof using standa shown the 7.2/me ium acetate
Abso

Eluent thanol
1 B: Methan (95:5, , pH
synthetic bonea-tions ter, P=0) 140
rd (i.e. 9 Sample al/Milli v/v)

rifi of DMT
oligonucleot 7 : -Q® water
modifi or Phosph

Antisense cation of Ph Details: P=S) oligon ides. ed (i.e. phospphosphole back- 1 = AMQ 12 protecte
orothioa
ted 21 (95:5,
2 = Asp 7 = NH3 Injectio mer (DMT v/v)
the produ ucleotides to horo-thioated 3 = Ser 8 = Arg 13 = Val n: 12 d)
4 mLs, contain

osphorot selectively 54=,=Gly 9 = Thr 14 = Met

Synthe ction of Glu

Therape tic oligon cells. As 10 = Ala 15 = Lys units (21 ing 624
targete inhibit
6 = His 11 = Pro 2 mgs) of O.D.26
used in ucleotides such, “antise d protein 16 = IIe 3

hio
13 in 70% oligonu 0nm

utic Inv techno 12 = Tyr 10

a s within 17 = Leu A and

estigatio ated DNA for

11 cleotide
ranging wide variety are succes
logy investi nse and 18 = Phe 30% B,
nse

17
Objective: from use of to multigram s-fully gations”
applic related 18 injected
into
e Respo

utilization as hybrid DNA. a-tions amoun require

milligram 0
and the as primers for rapid ization ts of purifie
ns Revers 2 15 Column
The object probe ed-pha d synthe 4 :
mV

polymerase Bondap
DNA seque and cost s to se HPLC
scenc

ive of 6
provides tic
demon Most ak® HC18
strate the this application recently,
4 8
reaction of ncingeffective techni
chainisolatio Eluent Radial-P
a
10
A: ak™ cartridgHA (8 x 100mm
Fluore

HPLC for utility of effectinote studies n* either

(PCR). que Milligra Minutes
12
14 100mM
reversed-pha venessis to havephosp horothioatedphosphodies for the 2 HC18 m quantit ammon e)
protected large scale purific phosp shown 9 16
of using the HA reverseies of DMT protect Eluent 18 7.2/me ium acetate
synthetic a-tions hodiesse ter, standa
oligon
rd (i.e.
ucleot full length ter or d-phas
B:
Methan20
thanol
(95:5, , pH
oligonucleot bone modifiof DMT P=0) or phosp applications ide products e materia ed synthetic Sample
: al/Milli
-Q® water
v/v)
ed (i.e. l in an oligonu 16 Phosph
Details: P=S)ides.
oligon phosp hole back-
2.4 . for these 1 3 7
8 8 x 100mm12 cleotid
es can protecte
orothio
ated 21 (95:5,
mer (DMT v/v)
6 Injectio
the produ ucleotides to horo-thioated
11 Radial- be rapidly d)
5 Pak n:
ction of selectively , 0 purified 12 mLs, contain
Synthe cells. from undesir ing
tic targete DMT blocked units
used in oligonucleotides techno As such, “antise d protein inhibit (21 mgs) ed 624 O.D.26
Wate rs Appl a
ranging wide variety are succes nse and s within
10 full length
product in 70%
A and
synthes
of oligonu is reactio
cleotide
0nm

icatio ns
logy investi 14 30% B, n by-pro
to multig s-fully related ducts using
Note s
from use of applic gations” Column injected
utilization as hybrid DNA.a-tions ram amoun require : Bondap
into Bondap
m

ts of purifie milligram 20 ak
and the as primers for ization Revers ak® HC18
e, 300n

probesed-pha Eluent Radial-P

se HPLC d synthetic
polymerase DNA rapid and Milligr Minutes
A: ak™ cartridgHA (8 x 100mm
cost to
am quantit
Most recent sequencing provides HC18 100mM e)
chain isolati
on of eithereffective techni HA reverseies of DMT protect ammon
rbanc

effectiveness ly, studies have reactio n* que for

a purifica d-phas Eluent 7.2/me 30
thanol
ium acetate
phosphoroth
shown
(PCR). phosp
the Detrityla tions as e materia ed synthetic Sample
B:
Methan (95:5, , pH
phosphodies of using standaoligon hodies v/v)
Abso

ted failures
and protectin require l oligonu al/Milli
the ioated full ter or g groups
in
d for antisen an 8 x cleotid : Phosph -Q® water
bone modifi ter, P=0) or rd (i.e.ucleot length se therape 100mm Radial- es can Injectio orothio (95:5,
applications ide products
phosp utic investig
be
Pak cartrid rapidly
protecte
n: purified d)
ated 21
mer (DMT v/v)
P=S) oligon ed (i.e. phosp hole back- . for these ations. ge. This 12 mLs, from contain
same undesi
the produ ucleotides to horo-thioated units packin
(21 mgs) ing 624
red synthes
g isof availab O.D.26 is
selectively ,
cells. As ction of targete
in 70% oligonule 0nm reactio
A and in a variety
cleotide n by-prod
d protein inhibit
0
30% B,
techno
such, “antise DMT blocked
injected of column ucts using Bonda
logy investi nse and s within 0 failures into dimens pak
to multig related 5 ion for gram
gations” 10
scale
ram amoun require 15
DNA. milligram 20
Reversed-pha ts of purifie Milligr 25
rapid and se HPLC d synthetic
Minutes 30
am quantit Moffat,
isolation cost effective HC18 35
provid Anti-Sen A.S. 1988
HA reverseies of DMT protect 40
of technique es a purifica Genetic se Technolo (Nov/Dec).
phosphoroth either phosp for the tions as d-phas
e materia ed synthetic Klauser. Engineering gy In Quest “Researchers
Bio/Tech A. 1990 News. for Novel Pursue
hodies require
d for antisenl in an 8 x
oligonu nology. . “Antisen Vol 8. Pages Drugs and
oligonucleot ioated full ter or cleotid se
se therape 100mm Radial- es can be rapidly
Vol 8. 1 and Agriprod
length Pages Start-Ups Surveyed 7. ucts”,
applications ide products utic investig Pak cartrid purified
303 - 304.
”,
. for these ations. ge. This from undesi
same packin red
g is availab synthesis reactio
le in a n by-prod
variety
of column ucts using Bonda

Wa t e r s
dimens pak
ion for
gram scale

Moffat,
Anti-Sen A.S. 1988
Genetic se Technolo (Nov/Dec).
Klauser. Engineering gy In Quest “Researchers
Bio/Tech A. 1990 News. for Novel Pursue
nology. . “Antisen Vol 8. Pages Drugs and
Vol 8. se 1 and Agriprod
Pages Start-Ups Surveyed 7. ucts”,
303 - 304.
”,

APPLICATIO
NS
NOTES

For your free Application Notebook File

Folder, please check box 7 on the Business
Reply Card and return today.

15
A New Solid Phase Extraction Method for
Determination of Triazine Herbicides and
Metabolites in Aqueous Samples
Michael S. Young
Introduction
Atrazine and related triazine herbicides
are used in great quantities throughout
the world for preemergence weed con-
trol. In the central United States, for
example, millions of kilograms of tri-
azines are applied each year. In areas
of heavy usage, surface water supplies
are often affected by runoff of these
substances and their transformation
products (1). Therefore, a number of
these compounds are included in the
list of drinking water contaminants to be
monitored under the Safe Drinking
Water Act. There is also significant
interest on the part of other agencies,
such as the USGS, regarding the fate
and transport of the triazine herbicides
in the natural environment. In Europe,
where groundwater is utilised for a high
proportion of drinking water supplies,
the EC has established more stringent
limits than has the US EPA. Currently,
the US limit is 3 g/L for atrazine; the
European limit is 0.1 g/L for atrazine
or any individual regulated pesticide,
and 0.5 g/mL for the sum of all pesti-
cides. Because groundwater levels in
agricultural areas were consistently
above this limit, Germany banned the
use of Atrazine in 1991, and has rec-
ommended banning the use of this her-
bicide throughout the European
Community (EC). Clearly, a rugged
method for determination of the triazine
herbicides is desirable with detection
limits in the part per trillion range.
Because direct determination at these
levels is not possible, sample enrich-
ment techniques, such as solid phase
extraction (SPE), must be employed. In
this study, Porapak® RDX Sep-Pak car-
tridges were used for trace enrichment
of triazines and metabolites.
Compared with traditional C18 silica or
styrene divinylbenzene polymeric phas-
es, the Porapak RDX material utilised in
this study shows enhanced retention
for organic compounds with polar
functionalities. Among the more polar
pesticides, atrazine and simazine are
compounds now routinely measured in
drinking water according to EPA
methods 507 and 505 (2). These
compounds are included in the class
of triazine herbicides. Often it is impor-
tant not only to measure a particular
analyte such as atrazine, but also the
biological transformation products of
that substance. For example, a number
of metabolites are of interest to
researchers studying the effects of wide-
spread atrazine use in the Mississippi
watershed (1,3). Figure 1 shows
atrazine and its principle transformation
products of interest to environmental
researchers. Atrazine and its sister tri-
azine compounds, (simazine,
propazine) are easily extracted from
water using C18 silica SPE procedures,
Figure 1:
(C2H5)NH N NH2(C3H7)
N N
Cl
ATRAZINE
N N
HN(C2H5) NH(C3H7)
N N
OH
HYDROXYATRAZINE
16
but the more polar metabolites, such as
desisopropylatrazine, are not suitable
for extraction using C18 silica.
Sep-Pak® cartridges containing Porapak
RDX resin were first developed for the
high efficiency extraction of nitramine
explosives from water (4). In this
research, the Porapak RDX cartridge
has been shown to be a superior SPE
device for determination of triazine
herbicides and metabolites in aqueous
samples. This method has been devel-
oped for use with natural waters
(surface and groundwater) as well as
finished waters (treated municipal
drinking water supplies). Samples (up
to 1.0 L) are adjusted to pH 7 with
phosphate buffer and are eluted
through the Sep-Pak cartridge at 10
mL/min. After extraction, the cartridge
is eluted using 1:1 methanol/acetoni-
trile and the resulting extract is analysed
by HPLC/UV using a Symmetry C18,
5µm column.
NH2 N NH2(C3H7)
N N
Cl
DESETHYLATRAZINE
C2H5HN NH2
N N
CI
DESISOPROPYLATRTAZINE
Experimental Section
Chemical Standards. Standard triazine
herbicides and metabolites were
obtained from AccuStandard Inc.
(New Haven CT). The analytes included
in this study were the herbicides
atrazine, propazine, cyanazine, and
simazine, and the metabolites desethy-
latrazine, desisopropylatrazine and
hydroxyatrazine.
Sources of Water. Detection limit studies
were conducted on high purity reagent
water obtained from a MilliQ ® system
(Millipore). Natural pond water was
obtained from a small manmade lake in
central Massachusetts. Finished drinking
water was obtained at Waters
Corporation, Milford, MA.
HPLC Conditions. All analyses were
performed on a Waters HPLC system
comprised of a 600 Multi-solvent
Delivery System, a 486 Tunable
Absorbance Detector, a 712 WISP ™
Autosampler, and a 616 LC system.
The Millenium® 2010 Chromatography
Manager was used for system control,
data acquisition, and data analysis.
The analytical column was a Waters
Symmetry C18, 5µm 3.9 mm x 150 mm.
The aqueous portion of the mobile
phase (Solvent A) was a pH 4.7 buffer
prepared from ammonium acetate and
acetic acid, each at 1 mM concentra-
tion in MilliQ water, and adjusted to
5% methanol. The organic portion of
the mobile phase (Solvent B) was 5%
methanol in acetonitrile. All solvents
were obtained from J.T. Baker
(Philipsburg N.J.) and were HPLC
grade; all reagents were also from
Baker and were ACS grade.
The injection volume was 50 L and
the detector was set to 223 nm for all
analytes with the exception of hydroxy-
atrazine, which was monitored at
240 nm.
Solid Phase Extraction with Porapak RDX
Cartridges. The conditions chosen for
the extraction of triazine herbicides and
metabolites were based on the proce-
dure developed for analysis of nitroaro-
matic and nitramine explosives (4). All
cartridge conditioning and sample
enrichment was performed at 10 mL per
minute. A Waters SPE manifold was
used for processing up to 12 cartridges
at a time. The sample size for finished
water was 500-1000 mL while the
sample size for non-filtered surface
water was 250 mL.
Recovery and detection limit samples
were first adjusted to pH 7 by addition
of 2 mL of 10-3 M phosphate buffer.
The Porapak RDX Sep-Pak cartridges
were then preconditioned with 15 mL
of methanol followed by 30 mL of
MilliQ water. The buffered samples
were then processed through the Sep-
Pak cartridges at approximately 10 mL
per minute. Following sample enrich-
ment, the cartridges were air dried for
10 minutes before elution with 5 mL of
1:1 acetonitrile/methanol. The sample
extracts were then diluted 1:1 with
MilliQ water before analysis by HPLC.
Results and Discussion
Effect of pH. Two separate experiments
were performed which showed little or
no relation between recovery and pH
in the range from 2 to 8.5. Preliminary
experiments have shown that the
Porapak RDX cartridge is highly effective
for extraction of phenolics at pH 2.
Therefore, the Porapak RDX cartridge
may be used for the simultaneous
extraction of triazines and metabolites
along with acidic contaminants such as
phenol.
Detection Limit. According to a proce-
dure recommended by the EPA (5), the
detection limit for an analyte may be
estimated from the magnitude of the
standard deviation in the results
obtained from seven or more replicate
spiked samples. In this experiment,
seven identical 1 L samples were
spiked at a level of 0.400 g/L of all
Table 1 Recovery of Triazine Herbicides and Metabolites from a Pond Water Source
(spike level 2 g/mL, 5 replicates)
Compound % Recovery
desisopropylatrazine 106
hydroxyatrazine 110
desethylatrazine 100
simazine 106
cyanazine 105
atrazine 107
propazine 103
17
components and were analysed by the
new Porapak RDX Sep-Pak method. The
results of this experiment indicated that
the detection limit for these compounds
is lower than 100 ng/mL.
Best Solvent for Sample Elution. Initial
experiments were performed using pure
acetonitrile to elute the analytes from
the Sep-Pak cartridges. However,
inconsistent recovery of hydroxyatrazine
led to an investigation of various sol-
vents and mixtures to yield consistent
and high recovery of all compounds.
It was observed that the addition of a
few drops of phosphoric acid to the
acetonitrile gave high recovery of the
hydroxyatrazine, but this procedure
resulted in an extract which required
pH adjustment before HPLC analysis.
An alternative elution solvent, 1:1
acetonitrile/methanol gave virtually
the same recovery as the acidified elu-
ent and required no further adjustments
of the extract for chromatographic
analysis.
Recovery from Pond Water. Four
replicate 250 mL samples were pre-
pared from the Lake Boon water
source. This was a typical surface
water with considerable particulate
matter, biota, and a noticeable green-
ish brown hue of dissolved and sus-
pended natural organic matter. The
results of this experiment are presented
in Table 1. These results confirm that
triazine herbicides and their principle
metabolites, including hydroxyatrazine,
are well recovered from a difficult
sample matrix.
% RSD
2.8
11
4.3
4.4
2.5
5.4
1.1
Figure 2 shows a typical HPLC chro- Figure 2: HPLC Analysis of Spiked Pond water Processed on Porapak RDX Sep-Pak Cartridge
matogram obtained from the spiked
pond water experiments. Column: Symmetry C18, 5µm 3.9 x 150 mm
Recovery from a Municipal Drinking Mobile Phase: A) 5% methanol in acetonitrile
Gradient: 100% A for 0.5 min step to 20% B,
Water. Five replicate 1.0 L samples hold for 5 min, then linear gradient to 75% B, for 25 min
were prepared at a spike level of 0.4 0.040 Flow Rate: 1.2 mL/min
g/mL using Milford, MA tap water. Injection Volume: 50 µL
0.035 Detection: 223 nm (240 nm for hydroxyatrazine)
Before spiking, each sample was treat-
Sample: 250 mL of pond water spiked
ed with 30mg of sodium sulphite to

8.150 deisopropyl atrazine

0.030 @ 2µg/L of each analyte
remove any residual chlorinating agent.
The results are presented in Table 2. 0.025

11.858 deethylatrazine
The high RSD value for recovery of 0.020

9.575 hydroxyatrazine

16.392 simazine
17.158 cyanazine
desisopropylatrazine resulted from

23.142 propazine
19.908 atrazine
0.015
one data point out of range. If that
data point is eliminated, the result for 0.010
recovery of that analyte is 101% with

18.867

20.400
an RSD of 6.5%. 0.005

Other solid phase extraction (SPE) 0.000

methods have shown acceptable recov-
ery for hydroxyatrazine but there are - 0.005
significant drawbacks to these proce- 10.00 15.00 20.00 25.00
dures. One porous polymer method Minutes
requires 3 hours to process a typical
sample. Another method, based on
The gradient chosen for this analysis was 100% A for 0.5 minutes, stepped to 20% B and
graphitised carbon black, requires a held for 5 minutes, and then a linear gradient to 75% B for 25 minutes.
more complicated and time consuming
solvent elution step. The Porapak RDX
Sep-Pak cartridge method described
here is the first SPE method which is Table 2: Recovery of Triazine Herbicides and Metabolites from a Municipal Drinking Water
Source (spike level 0.4 g/mL, 5 replicates)
fast, accurate and reproducible for the
determination of triazine herbicides and Compound %Recovery % RSD
metabolites in water.
desisopropylatrazine 100 6.5(n = 4)
Conclusion hydroxyatrazine 86.0 4.9
The Porapak RDX Sep-Pak cartridge has desethylatrazine 103 8.1
been shown to be a highly effective simazine 95.5 3.4
solid phase extraction device for the cyanazine 98.0 7.7
determination of triazine herbicides and atrazine 97.0 5.6
metabolites in aqueous samples. propazine 104 2.2
Recovery was greater than 95% for
atrazine, simazine, cyanazine and
propazine, and greater than 90% for
desethylatrazine, desisopropylatrazine,
and hydroxyatrazine. The high recovery References
of hydroxyatrazine is of great impor- 1. W. Pereira and C. Rostad, Environ. Sci. Technol. 24 4. E. Bouvier, L. Steriti, and S. Oehrle, 23rd Annual
tance to researchers and investigators (1990), 1400 Northeast Regional Meeting of ACS, Boston MA, 1993
interested in the fate and transport of 2. “Methods for the Determination of Organic Compounds 5. ”Definition and Procedure for the Determination of the
triazine herbicides. in Finished Drinking Water” (1988), EPA-600/4-88/039 Method Detection Limit”, 40 CFR 136, App. B.
(1991 revision)
The Symmetry C18 column employed for
3. E. Thurman, D. Goolsby, M. Meyer, and D. Kolpin,
this study proved to be rugged and reli- Environ. Sci. Technol. 25 (1991), 1794
able. Although some performance loss
was noted at the end of the study, after
some 300 injections in various matri-
ces, the column performance was still
acceptable for sub ppb analysis.

18
Ordering Information
Porapak RDX

Item Part Number

fi Herbicides
Accurate analysis of herbicides
following solid phase extraction and
Porapak RDX Cartridges WAT047220 concentration requires high sensitivity
(box of 30)
detection technology. Waters 486 Tunable
Nova-Pak C8 WAT035876 Absorbance Detector is the highest
3.9 x150 mm column
performing UV/Vis detector with the
Nova-Pak C8 WAT052805 flexibility to accommodate a wide range of
3.9 x100 mm cartridge column
HPLC methods. Waters exclusive Taper-Cell ®
Nova-Pak CN HP WAT052795 flow cell eliminates refractive index effects,
3.9 x100 mm cartridge column
while a dual-beam optical design reduces
End connector kit WAT037525 noise resulting in superior signal-to-noise
(required for cartridge column)
performance. Programmable wavelength
Waters Vacuum Manifold, WAT054750 switching, accomplished with ±0.25 nm
12 position
precision using digital stepper motor tech-
Waters Vacuum Manifold, WAT054755 nology, enables automatic selection to
16 position
further optimise detection sensitivity.
Waters Vacuum Manifold, WAT054760
24 Position
Tubing, Tefzel,1/8inch WAT023344
O.D. x 0.040-inch I.D
Sep-Pak Vac Adapter WAT054260
(Box of 12)
60 cc Sep-Pak Reservoir WAT024659
(Box of 12)
For more information on

Porapak RDX please

check box 9 on the Business
Reply Card and return today

For information on the Compendium of Technical

Explosives in surface and groundwater, please
check box 9 on the Business Reply Card
and return today.

For more information about Volume 6 of the

Solid Phase Extraction Applications Guide and
Bibliography, please check box 8 on the Business
Reply Card and return today.

19
Measurement of Trace Formaldehyde in
Workplace Air using Waters XPoSure
Aldehyde Sampler Cartridges
Donna M. Martin and Pamela C. Iraneta
A Short Term Exposure Limit (STEL)
method for the analysis of formaldehyde
in workplace air was developed using
Waters XPoSure™ Aldehyde Sampler
Cartridges. The method involves collect-
ing air onto a cartridge that contains a
solid phase derivatising agent specific
for aldehydes and ketones, eluting the
sample and analysing it by HPLC. The
high flow rate capability and guaran-
teed low backgrounds allow for the
quantitation of formaldehyde down to
the 0.02 ppm range. Results from
method validation studies demonstrate
that XPoSure meets the National Institute
for Occupational Safety and Health
(NIOSH) accuracy criterion.
Introduction
Formaldehyde is receiving increased
attention as a toxic substance and an
occupational hazard. It is widely used
as a textile treatment agent, as an inter-
mediate in the manufacture of urea-
formaldehyde, and in making phenol-
formaldehyde resins which are often
used in manufacturing fiberboard, parti-
cleboard and plywood.
The Occupational Safety and Health
Administration (OSHA) reduced the 8
hour permissible exposure limit (PEL)
for formaldehyde from 1 ppm to 0.75
ppm and the American Conference of
Governmental Hygienists (ACGIH) has
adopted a new threshold limit value
(TLV) of 0.3 ppm.(1,2) This trend
toward lower exposure limits is expect-
ed to continue. In particular, Short Term
Exposure Limit (STEL) measurements for
formaldehyde in the sub ppm range
have been difficult to quantify with
certainty using older solid phase
derivatisation technologies such as 2-
(hydroxymethyl) piperidine (3). The
difficulty is due to the short sampling • The method needs to quantify
time of the method and the background formaldehyde in the range of 0.02
levels of the derivatisers. In the STEL to 2 ppm under low and high
method, air samples are taken for 15 humidity conditions.
minutes with portable sampling pumps • The method needs to meet the
ideally placed in the breathing zone. National Institute for Occupational
The Waters team discussed STEL mea- Safety and Health accuracy
surements for formaldehyde in work- criterion.
place air with several experts in industri-
A cutaway view of the XPoSure
al hygiene and occupational safety. Aldehyde Sampler cartridge is shown
The following is a list of performance
in Figure 1. It has male and female luer
criteria for the ideal sampling device: connectors for easy attachment to
• The solid phase derivatisation pumps and is bi-directional with respect
reaction must be fast. to sample collection and desorption.
• The device needs to be compatible The solid phase is 2,4-dinitrophenylhy-
with existing portable sampling drazine (DNPH) coated silica which,
pumps used in workplace under acidic conditions, reacts rapidly
monitoring. with aldehydes and ketones to form its
• The device needs to accommodate
a large sample throughput to pre
concentrate formaldehyde on the
cartridge.
Figure 1: Cutaway View of an XPoSure Aldehyde Sampler Cartridge.
Female luer connector
Polyethylene frit 100 µm pore size
Acidified DNPH coated silica
500 -1000 µm particle size
Polyethlene frit 100µm pore size
Aluminum compression ring
Male luer connector
20
corresponding DNPH-hydrazone deriv-
ative (shown below). The DNPH-hydra- NO2 NO2
zone can then be eluted from the car-
R1 H+ R1
tridge with acetonitrile and analysed C=O H2NNH NO2 C = NNH NO2 + H2O
by HPLC. R R

Since 1991, Waters has marketed a Carbonyl Compound 2,4-Dinitrophenylhydrazine DNPH Derivative
DNPH-coated product that measures (Aldehyde or Ketone) (DNPH)
trace aldehydes and ketones in car
exhaust for the automotive industry.
XPoSure is similar in design, however,
a larger silica particle size and larger Figure 2: Chromatographic Separation used to Analyse Samples
frit pore size are used to obtain low
back pressure and high sample
Column: Symmetry™C18 5 µm 3.9 x 150 mm
throughput compatible with portable Mobile Phase: 45:55 acetonitrile/water
sampling pumps. Flow Rate: 1.3 mL/min
Detection: 360nm
Injection Volume: 20 µL
Experimental Calibration Standard Level 4

Sampling and Analysis: Air with known 1. DNPH 2.5 µg/mL

0.10 2. DNPH-Formaldehyde 0.29 µg/mL as Formaldehyde
amounts of formaldehyde was sampled 3. DNPH-Acetaldehyde 0.39 µg/mL as Acetaldehyde
2
for 15 minutes at a flow rate of 1.5 4. DNPH-Acetone 0.49 µg/mL as Acetone
1
L/min (22.5 L of air total). The car- 0.08
3
tridge was then capped and sent to the
lab for analysis. The DNPH-hydrazone 0.06
4
derivatives are eluted with acetonitrile
AU

at a flow rate of 3 mL/min into a 10

0.04
mL volumetric flask. 20 µL of the eluate
was injected onto the HPLC.
0.02
Fraction collection experiments show
that this method consistently provides
recoveries of greater than 98 percent(4). 0.00
Stability data show that collected 0 2 4 6 8 10 12 14
samples can be stored for up to two Minutes

weeks at 4 °C or up to 4 days at room

temperature.(4)
The HPLC chromatogram for the
DNPH-hydrazone derivatives of:
formaldehyde, acetaldehyde and
acetone is shown in Figure 2. A
Waters Symmetry® C18 5µm column
(3.9 x 150 mm) was used under
isocratic conditions of 45% acetonitrile
in water mobile phase. The Waters™
HPLC system consisted of a 600
Multi-Solvent Delivery System, a 486
Tunable Absorbance Detector, and a
715 Ultra WISP Sample Processor.
The absorbance was monitored at
360 nm. The data were collected and
processed using Waters ExpertEase
Chromatography Software Version 3.2
installed on an 845 Workstation.
Method Validation: Accuracy and
precision were tested using controlled
atmospheres of formaldehyde at
2 ppm, 1 ppm, 0.5 ppm and
0.02 ppm for two relative humidity the water and reached saturation at the
conditions of <10% and >85%. A outlet of the bottle. This provided a RH
schematic of the experimental set-up of >85 % at the sampler. The natural
is shown in Figure 3. A 15 cm humidity of the system was <10 %.
Dynacal permeation tube holder and The temperature of the bath was varied
heat exchanger (VICI Metronics, between 72 to 100 °C depending on
Santa Clara, CA) was placed in a the concentration of formaldehyde
controlled temperature silicone oil bath. desired. The permeation rate was
Formaldehyde at 2 ppm, 1 ppm, and determined by monitoring weight loss
0.5 ppm was generated using an 8 cm of the permeation tubes over time.
high emission Dynacal permeation tube Actual formaldehyde concentration
and formaldehyde at 0.02 ppm was was determined by dividing the
generated using a 3.0 cm standard permeation rate by the flow rate.
permeation tube (VICI Metronics).
Nitrogen was the carrier gas. Humidity
was introduced into the system by plac-
ing a 2 L bottle of water in the side
arm. Nitrogen was sparged through

21
Figure 3: Schematics for Permeation Tube Systems used to Generate Atmospheres of Known Formaldehyde Concentrations

0.2 L/min

Flow Meter Sampler Flow Meter

N2

Permeation Tube

1.3 L/min

0.2 L/min

Flow Meter Sampler Flow Meter

N2

Permeation Tube

1.3 L/min

H2 O

Results and Discussion

The precision and bias data for 10 to
12 replicate measurements at >85%
and <10% relative humidity are shown
in Figure 4. The relative standard devia-
tions ranged from 1.99 to 4.61. The
pooled relative standard deviation was
found to be 3.39. Assuming an addi-
tional 5% relative standard deviation
from sampling pumps, the estimated
total relative standard deviation for the
STEL method is 6.4.
The method bias was determined by
comparing the HPLC results of the
XPoSure Aldehyde Sampler Cartridge
with the permeation tube weight loss
data. There is excellent agreement with
a bias range from 0.4 to 8.3 at lower
concentrations. The increase in bias at
the lower concentration range is proba-
bly due to the difficulty in measuring
small weight losses at lower permeation
rates. The accuracy criterion defined
by NIOSH is a function of the bias
and precision. It states that a method
must give results that are within ±25%
of the true value at a probability level
of 95%.(5) The confidence statistics are
well within this range and the method
described here meets the NIOSH
accuracy criterion.

22
Figure 4: Precision and Bias Data for 15 Minute STEL Measurements of Laboratory Generated Controlled Atmospheres Containing
Formaldehyde

Nominal Replicates Relative Sampler Sampler Wt. Loss Biasb

[CH2O] Humidity Result a RDS Result b %
n ppmv ppmv

2 ppmv 10 <10% 1.97 2.17 1.93 2.1

12 >85% 2.03 1.99 2.02 0.4

1 ppmv 12 <10% 0.93 4.61 0.94 - 0.7

12 >85% 0.95 3.00 0.94 0.9

0.5 ppmv 12 <10% 0.41 3.57 0.43 - 4.5

12 >85% 0.43 2.85 0.43 - 0.4

0.02 ppmv 12 <10% 0.020 4.00 0.020 4.2

12 >85% 0.021 3.85 0.020 8.3

a Samples were obtained using a sampling flow rate of 1.5 L/min.

b Bias was determined by comparison to permeation tube weight loss (Wt. Loss).

Ordering Information
XPoSure

Quantity Part No.

XPoSure Box of 20 WAT047205 For more information on

Aldehyde Sampler Cartridge
Symmetry C18, 5 µm column, One WAT046980 XPoSure Aldehyde Sampler
3.9 x 150 mm Cartridge, please check box 10
End Connector Kit One Kit WAT037525 on the Business Reply Card
and return today.

Reference
1. U.S. Department of Labor, Occupational Safety and Cincinnati, OH: NIOSH 1984. Method 2541-
Health Administration (OSHA): Code of Federal Regulations Formaldehyde.
Title 29 CRF 1919.1048 Formaldehyde, Government
Printing Office, Washington, D.C., July 1, 1993. 4. P. C. Iraneta, presentation at AICHE 1995, Kansas City,
pp. 410-442 May 1995

2. American Conference of Industrial Hygienists (ACGIH): 5. E.R. Kennedy, T.J. Fischbach, R. Song, P.M. Eller, and S.
1991 – 1992 Threshold Limit Values and Biological A. Shulman, “Guidelines for Air Sampling and Analytical
Exposure Indices. Cincinnati, OH: ACGIH, 1991. Method Development and Evaluation”. Cincinnati, OH:
NIOSH July 13, 1994. p.38.
3. National Institute for Occupational Safety and Health
Administration (OSHA): NIOSH Manual of Analytical
Methods, 3rd ed. (DHHS/NIOSH Pub. No. 84-100)

23
Literature Corner:
Clinical:
Andrew Clifford and his colleagues at UCal/Davis have developed an innovative
“Protocol for Collection and HPLC application for Waters Sep-Pak ® DNPH-Silica Cartridges. They collect normal human
Analysis of Volatile Carbonyl breath over a 15-20 minute period, draw a 25-litre aliquot of this sample through a
Compounds in Breath” DNPH-Silica Cartridge, elute the hydrazone derivatives of trapped aldehydes and
ketones with 3 mL of acetonitrile, and identify the compounds by HPLC analysis.
Y. Lin, S.R. Dueker, A.D. Jones, S.E. Ebeler, Recoveries were typically >90%, and CVs for the entire sampling/analysis protocol on
and A.J. Clifford, Clin. Chem. 41(7) duplicate samples were <11%. This new protocol is noninvasive and will enable many
(1995), 1028. in vivo studies on the links between levels of particular carbonyl compounds in human
breath and various metabolic or pathological states, for example, acetone (diabetes
mellitus), acetaldehyde (alcohol consumption.)
Patrick McDonald

Industrial/Environmental:
Peter Jackson and collaborators have demonstrated a variety of ways in which HPLC
“Studies on System Performance and systems can be configured with combinations of hardware, suppressor, and column
Sensitivity in Ion Chromatography” components for anion analysis. Dual-piston, reciprocating pumps fitted with low-pressure
pulse dampeners minimised baseline noise seen with conductivity detection, whether
P.E. Jackson, J.P. Romano, and B.J. or not post-column suppression was used. Column temperature control was critical to
Wildman, J. Chromatogr. A 706(1+2) successful sub-ppb detection limits in IC without suppression. An optimum system for
(1995), 3. selectivity as well as detection limits was the Waters IC-Pak™ Anion HR column (4.6 x
75 mm) with a Waters IC-Pak Cation Guard Column (4.6 x 50 mm) used as a
suppressor. Though solid-phase suppressors have a finite lifetime, before regeneration is
required, high-capacity, well-packed suppressor columns like the IC-Pak C column
demonstrate an optimum combination of sensitivity, linearity, and analytical efficiency.
Patrick McDonald

fl fl fl fl fl fl fl fl fl fl fl fl fl fl fl fl fl fl fl fl fl fl fl fl fl fl fl fl fl fl fl fl fl fl fl fl

On-line and off-line solid-phase (SPE) extraction methods were developed for
“Degradation of Organophosphorous determination of organophosphorous insecticides and metabolites in natural waters.
Pesticides and their Transformation The methods were used to investigate the degradation rates of these materials in the
Products in Estuarine Waters” environment. Following SPE on C18 silica, separation and detection was performed
using a number of techniques including HPLC coupled with photodiode array detection
S. LaCorte, S. Lartiges, P. Garrigues, and (PDA) and thermospray mass spectrometry (TSMS). The authors observed that the on-line
D. Barcelo, Environmental Science & HPLC based methods employing sensitive detectors such as the Waters 996 Photodiode
Technology 29, no.2 (1995), pp. 431-438 Array Detector were the most sensitive of the techniques investigated, allowing detection
limits of 0.1 g/L using only 150 mL of sample. They also obtained these detection limits
for analytes which are too thermally labile or polar for traditional GC based analysis.
Based on their analyses, the authors have concluded that many of the insecticides are
more persistent in the environment than previously thought.
Michael S. Young

Biochemistry:
“Synthesis and Purification of Milligram
Quantities of Short RNA Transcripts”
R. Kim, E. L. Holbrook, J. Jancarik and S.
Kim, BioTechniques 18(6) (1995), 992
Purification of milligram amounts of RNA is necessary for biochemical and biophysical
studies, such as x-ray crystallography and NMR. The authors describe a one-column
method that uses a 90 min step gradient to separate runoff RNA transcript molecules
that differ by one nucleotide (nt) out of 69 nt. The isolated material is pure enough to
be used for the biophysical studies. A double stranded DNA template made by
polymerase chain reaction (PCR) was used to synthesise the RNA chains by in vitro
transcription. The RNA was precipitated, solubilised and then fractionated on a
DEAE-5PW ion-exchange column (Waters Protein-Pak™ DEAE-5PW) on the Waters
650 Advanced Protein Purification system. The 70 nt was separated from the 69 nt.
All unincorporated nucleotides elute early in the low salt buffer and any aggregates
elute in the high salt buffer. The authors also state that the method has been used in
their laboratory to transcribe and purify several different RNA molecules ranging in
size from 32-to 69-nt long.
Dorothy J. Phillips

24
 Pharmaceutical:
Renin inhibitors are potentially better for controlling hypertension than the currently used
“Determination of the Renin Inhibitor
angiotensin converting enzyme (ACE) inhibitors. Renin inhibitors have the potential to
Ro 42-5892 in Human Plasma by produce fewer clinical side effects; therefore, the search for them has become an active
Automated Precolumn Derivitisation, research area. Ro 42-5892 has been identified as a very potent renin inhibitor in man
Reversed-Phase High Performance and required a sensitive assay method for the evaluation of its pharmacokinetics. The
Liquid Chromatographic Separation authors report a very sensitive and selective HPLC assay for analysis of Ro 42-5892
and Electrochemical Detection after in human plasma. The Ro 42-5892 was extracted from plasma with dichloromethane,
Post-column Irradiation” derivatised with 2,4-dinitrofiuorobenzene, chromatographed and then detected by elec-
J. Leube and G. Fisher, J. Chromatogr. B trochemical oxidation after UV irradiation. Identifying the stationary phase that gave the
665 (1995), 373
desired selectivity was a challenge. Fifteen different silica-based phases (3 to 10 µm
particle size, different chemistries) and three other phases were tested in 24 different
column formats. Only Waters Nova-Pak® C18 4 µm 3.9 mm x 150 mm column gave
the required selectivity between interfering peaks and the analyte. The mobile phase
contained pH 7 buffered acetic acid with disodium-EDTA and acetonitrile (100:85).
The combination of the excellent selectivity on the Waters Nova-Pak C18 column and
the electrochemical detection achieved a limit of quantitation of 0.3 ng/mL.
Quantitative recovery of Ro 42-5892 of 98% and 107% were noted at 0.5 and 3
ng/mL concentrations, respectively. The assay was used successfully for over 250
samples from a pharmacokinetic study in hypertensive patients.
Dorothy J. Phillips
fl fl fl fl fl fl fl fl fl fl fl fl fl fl fl fl fl fl fl fl fl fl fl fl fl fl fl fl fl fl fl fl fl fl fl fl

“Simultaneous microassay of alfentanil,
fentanil and sufentanil by high-perfor-
mance liquid chromatography”
R. Bansal and J.V. Aranda, J. Liq. Chrom.
18 (1995), 339.
Researchers at McGill University and Lady Davis Institute of Medical Research have
developed a rapid HPLC assay that simultaneously quantitates alfentanil, fentanil
and sufentanil in plasma and urine samples of newborn babies. A Nova-Pak CN 5µm
cartridge, 8 x 100 mm was used along with a guard column of the same packing.
Detection limits as well as linearity range were estimated. Due to the speed and
sensitivity of the assay (less than 10 minutes), it can be easily used for therapeutic
drug monitoring of these drugs.
Zoubair El Fallah

Food:
There is concern that organophosphorous pesticides, such as methyl parathion, may
“Rapid Determination of Methyl contaminate dairy animals through ingestion of contaminated grasses or by percuta-
Parathion and Methyl Paraoxon in Milk neous absorption. A rapid solid phase extraction (SPE) procedure has been developed,
by Gas Chromatography with Solid- utilising Waters Sep-Pak C18 cartridges, for the isolation and concentration of methyl
Phase Extraction and Flame parathion and methyl paraoxon residues from milk prior to GC analysis. When com-
Photometric Detection” pared with the traditional liquid-liquid extraction (LLE) procedure, the SPE procedure
gave similar detection limits, fewer interferences, and used far less solvent. A typical
R. Baynes and J. Bowen, Journal of
sample could be prepared by SPE in 75 minutes compared with over 3 hours for
AOAC International, 78 no 3 (1995),
the LLE procedure.
pp. 812-815 Michael S. Young

General Chemistry:
Because of its non destructive nature and ease of calibration, researchers at Rice
“ Determination of C60 and C70 University used HPLC with photodiode array detection to quantitate fullerenes in
Fullerenes in geologic materials geologic materials. A Waters Nova-Pak C18, 4 µm 3.9 x 300 mm column was used.
by high performance liquid The mobile phase was 40/60, toluene/methanol and quantitation was performed
chromatography” using a Waters 996 Photodiode Array Detector at 330 nm. Limits of detection of 5
and 20 ng were estimated for C60 and C70, respectively. Due to the high organics
D. Heymann, L.P. Felipe Chibante and R.E. content in real samples, a preparative step was necessary. It was performed on a Prep
Smalley, J. Chromatogr. A 689 (1995), Nova-Pak C18 6 µm 19 mm x 300 mm column. Collected fractions of C60 and C70
157. were then reinjected on the analytical column. Recoveries on the order of 90% were
obtained. Due to possible photochemical decomposition of early eluting organics,
which could contaminate the detector cell, the authors suggest running the preparative
step with the PDA turned off.
Zoubair El Fallah

25
Conversations with Chemists
Jeff Mazzeo Michael Swartz
In the last issue of the Waters Column,
we published an article on the
"Separation of Enantiomers
by Micellar Electrokinetic Capillary
Chromatography with Novel Chiral
Surfactants.” The response was over-
whelming, so we decided to follow-up
the article by interviewing the
Waters' scientists who developed
EnantioSelect™. EnantioSelect reagents
guarantee exact enantiomer migration
order reversal for improved quantita-
tion and peak identification.
If you would like to receive the last
issue of the Waters Column, please
check box 11 on the Business Reply
Card and return today.
Ed Grover
What role did you play in the
development of EnantioSelect?
Jeff Mazzeo: My initial role on the
chiral CE team was to evaluate the
ability of novel chiral surfactants pre-
pared by Ed Grover to separate enan-
tiomers via MEKC. Since many of the
chiral compounds which we are inter-
ested in separating are charged, I
developed equations to calculate the
capacity factors and alpha values of
charged compounds, and also exam-
ined the effect of migration window on
resolution for both neutral and charged
compounds. During initial evaluation, I
determined that optimum partitioning
was just as important as enantioselec-
tivity in determining resolution.
Specifically, partitioning of anionic
analytes was low due to charge repul-
sion. To overcome this problem, work
would have to be done at a pH below
the pKa of the acidic group in the ana-
lyte structure, which led Ed to synthe-
size anionic surfactants which main-
tained full ionisation at acidic pH’s.
Michael Swartz: I first investigated the
separation of chiral compounds using
CE in 1989 using cyclodextrins as
buffer additives. These early efforts
were frustrated by a lack of selectivity
and applicability--every new com-
pound required an intense application
effort that was too often unsuccessful.
Upon the realisation that synthesised
chiral surfactants solved many of the
frustrating problems associated with
previously investigated additives, I've
focused my energies working on creat-
ing method validation protocols for
EnantioSelect.
Ed Grover: During the research phase,
I was responsible for the systematic
design and synthesis of chiral surfac-
tants to understand the influence of
26
surfactant structure on the separation of
enantiomeric mixtures by CE. Over
eighty proprietary chiral surfactants
were synthesised by me, and their
propensity to separate enantiomers via
a MEKC format were tested by Jeff
Mazzeo and Mike Swartz. It has been
exciting to work on the separation of
enantiomeric mixtures by CE. Not only
is the study of chirality very relevant to
industry and the consumer marketplace,
it also provides an intriguing academic
challenge.
What is the potential impact of this
technology?
Ed Grover: If we are ultimately able to
develop a complementary set of chiral
surfactants (five to six surfactants) which
can separate most of the chiral com-
pounds in existence today, we believe
that we would greatly simplify the ana-
lytical chemist's approach to separating
enantiomeric mixtures.
Michael Swartz: Although the true
impact is yet to be seen, I think that we
have brought a new way of thinking
about the assay of chiral compounds to
the scientific community. While compar-
isons to other additives will be drawn,
the advantages of our approach--simul-
taneous achiral and chiral compound
analysis, peak reversals for identifica-
tion and quantitation, and simplified
method development--should not be
underestimated.
Jeff Mazzeo: The unique ability to
exactly reverse enantiomer migration
order will improve quantitation of enan-
tiomeric separation and permit confir-
mation of chiral separation. This benefit
is clearly seen in comparison to hetero-
geneous chemistry such as chiral LC
phases, as well as natural product CE
additives such as derivatised cyclodex-
trins. I believe that once we generate a
family of chiral CE additives which will
complement each other in terms of their
ability to separate chiral compounds,
and will share the benefits described
above, enantiophoresis technology will
revolutionise the way chiral separations
are performed.
What’s new
Welcome to Waters on the World Wide Web
Visit us on the internet at http://www.waters.com/
You may search the entire Waters site
to find the information of interest to
you. If you don’t find the Waters’
products and services you need, please
fill in our electronic form, send e-mail
to info@waters.com, or phone or
FAX your local Waters office. We’ll
be in touch “real-time” as quickly as
possible.
Waters Poster Presentations
at Pittcon 1996
• Chromatographic Characterization of a
Novel Octylcarbamate Bonded Phase.
• The Importance of Good Shape in the
Choice of a Preparative Packings for
the isolation of Drug Impurities and
Degradation Products.
• Faster Analysis with High Resolution
Using A New Small Particle Packings.
• Role of Narrow-Bore Columns in HPLC
Method Development.
• A New Porous Polymer Solid Phase
Extraction Cartridge for Determination of
Polar Analytes in Environmental Samples.
Waters News and Views
Includes corporation news, recently published
articles from Waters, Millennium 2010
Chromatography Manager and 2020
Client/Server Networks, Analytical Columns and
Sample Prep Devices,and Interactive Computer
Presentation Modules (CPM’s) from Waters, and more. For more information
contact: Kathy Fernandes, Product Publicity
Tel: (508) 482-2269, FAX: (508) 476-3470, e-mail: kathyf@tiac.net
Waters Chromatography On-Line Catalogs
Chromatography Chemistry Catalog
Chromatography Products for Sample Preparation,
Analysis, and Purification.
Waters Instruments Parts & Supplies Catalog
This catalog contains the largest inventory of quality parts and supplies to keep
your HPLC system operating at peak efficiency and validation levels
Waters On The Road. Trade Shows, Seminars and HPLC Schools
We provide you with the most up-to-date information on
trade shows, seminars, and HPLC Schools that Waters is
sponsoring or attending worldwide.
Looking for accurate and consistent results for analytical sample
filtration? Introducing high-performance Gelman Sciences
filters from Waters that give you a edge in HPLC.
WHAT GIVES YOU THE EDGE
When you purchase Gelman Sciences Acrodisc® syringe filters
and membrane filters from Waters you get sample and solvent
filtering that’s certified for compliance. Participation in both the
Waters Frequent Buyer Program and Gelman Sciences Frequent Filter
Program so you can earn points toward gifts, products
and training courses.
For more information on the Waters Edge Program, please check
box 12 on the Business Reply Card
27
How to reach us
Sales Offices Mexico Waters S.A. DE C.V.
Austria and European Export (Central Europe, Tel.: 525 359 0121
CIS, India and India Subcontinent) Fax: 525 359 0394
Waters Ges.m.b.H. The Netherlands Waters Chromatography B.V.
Tel.: 1 8771807 Tel.: 7650 87200
Fax: 1 8771808 Fax: 7650 87280
Australia Waters Australia Pty. Limited Norway Waters AS
Tel.: 2 933 1777 Tel.: 2 219 5360
Fax: 2 898 1455 Fax: 2 219 5362
Belgium and Luxembourg Waters S.A. - N.V. Peoples Republic of China - Beijing
Tel.: 02 7261000 Waters China Ltd.
Fax: 02 7261100 Tel.: 10 592 8918
Brazil Waters Comercial LTDA Fax: 10 506 0344
Tel.: 011 55 11 543 7788 Poland Waters Sp.zo.o
Fax: 011 55 11 530 6413 Tel.: 48 2 669 1225
Canada Waters Limited Fax: 48 2 663 7033
Tel.: 800 252 4752 Puerto Rico Waters
Fax: 905 678 9237 Tel.: 809 747 8445, 790 2225
CIS Waters Representation Office, Moscow Fax: 809 747 8448
Tel.: 7 095 336 7000 Singapore Waters Asia Ltd. (Asia Headquarters)
Fax: 7 095 336 7000 Tel.: 225 4855
Czech Republic Fax: 225 4655
Waters Ges.m.b.H. Spain Waters Cromatografia S.A.
Tel.: 42 2 35 02 27 Tel.: 3 325 9616
Fax: 42 2 35 23 75 Fax: 3 325 9896
Denmark Waters A/S Sweden Waters Sverige AB
Tel.: 46 598080 Tel.: 08 6282440
Fax: 46 598585 Fax: 08 6282455
Finland Waters Oy Switzerland Waters AG
Tel.: 0 506 4140 Tel.: 62 889 2030
Fax: 0 5064 1411 Fax: 62 889 2059
France Waters S.A. Taiwan Waters Asia Ltd.
Tel.: 1 30487200 Tel.: 2 706 8766
Fax: 1 30487201 Fax: 2 706 9704
Germany Waters GmbH UK and Ireland Waters Ltd.
Tel.: 06196 40 06 00 Tel.: 1 923 816700
Fax: 06196 48 23 88 Fax: 1 923 219012
Hong Kong Waters China Ltd.
Tel.: 2964 1800 All other countries
Fax: 2549 8956 Waters Corporation
Hungary Waters kft 34 Maple Street
Tel.: 1 1620686 Milford, MA 01757 USA
Fax: 1 1853236 Tel.: 508 478 2000
India Waters Pvt. Ltd. Toll-free: 1 800 252 4752
Tel.: 80 3341944 Fax: 508 872 1990
Fax: 80 3340527
Italy Waters S.p.A.
Tel.: 02 2500565
Fax: 02 2501827
Japan Nihon Waters Ltd.
Tel.: 3 3471 7191
Fax: 3 3471 7116
Malaysia Waters Asia Ltd.
Tel.: 3 704 8600
Fax: 3 704 8599

You can also visit us on the internet at http://www.waters.com/

28
Paraquat/Diquat

Objective:
The current EPA method for analysis of
aqueous samples for paraquat and diquat 360.00
is method 549.1. This method specifies a
340.00 Mobile phase: 70% methanol/30% water with
photodiode array detector (PDA) because the
320.00 PIC B-7, pH = 3.0
HPLC procedure chosen in the method does 308 nm
300.00

AU
not fully resolve the peaks. Paraquat is moni- Flow Rate: 1.0 mL/min
280.00
tored at 257 nm, while diquat is monitored at Column: Symmetry C18, 5 µm 3.9 mm x 50 mm
260.00 257 nm
308 nm. The EPA method used a C18 column Temperature: 30 °C
240.00
and a mobile phase containing diethylamine
220.00 Detection: 257 nm and 308 nm
(DEA), to maintain good peak shape. The
objective was to provide an improved chro- Data System: Millennium 2010
matographic separation without the need of Paraquat
Chromatography Manager, Ver. 2.01
0.040
an amine modifier.
Details: 0.030
Diquat
®
A Symmetry C18 5µm column was used for
AU

0.020
this analysis because it does not require any
0.010
amine modifier to prevent tailing. Also, its
high level of retention for the analytes allowed 0.000
the use of a 50 mm column. The short col- 6.00 7.00 8.00 9.00 10.00 11.00
umn, along with the elevated temperature uti- Minutes
lized, allowed the analysis to be accom-
plished in under 15 minutes.

System:
A Waters 600 Multi-Solvent Delivery System with a column heater was
used with a 717 Plus Autosampler and a 996 Photodiode Array
Detector. System Control, data acquisition and reporting were
accomplished with the Millenium 2010 Chromatography Manager,
Version 2.01.

Waters Corporation 34 Maple Street, Milford, Massachusetts 01757-3696 U.S.A. 508 478-2000
Millennium 2010 Chromatography Manager
Advanced computer technology simplifies accessing,
interpreting, and reporting your results.
The productivity of today’s chromatography is centered
around the ability to find, generate, summarise, and report
results in a way that meets your needs and complies with
any regulatory requirements. The Millenium relational data-
base lets you create your own view filters to customise the
way you sort your information, such as “lotCode”.
• Using the database, you can track a method history
giving you audit trail capability that ensures total security
of your raw data and results. In addition, you can
specify “Access Type” which allows you to “lock” your
methods for even greater security.
• The database keeps track of your instruments, triggering
an alarm when scheduled maintenance is required. The
Millenium relational database lets you view the injection
information on any calibration point at any particular
point in time on a processing sequence. This gives you
a detailed account of the processing method for that
injection, ensuring an accurate audit trail for regulatory
compliance.
• The Millenium Chromatography Manager can provide
acquisition and control for as many as four chromato-
graphic systems, including three HP5890 GC.
• The Millenium Report Publisher has the capability to give
you exactly what you want.
Waters 717plus Autosampler
Waters 717plus Autosampler is a fully programmable
autosampler for High-Pressure Liquid Chromatography.
Waters 996 Photodiode Array Detector
Complete Spectral Data-Complete Confidence • You can centralize operations through single-keyboard
in Peak Purity and Identity. system programming with either the new Waters 600E
controller or with an integral, on line PC.
With Waters 996 Photodiode Array Detector, you can be
confident that if there is an impurity in your sample, you will • For comprehensive programming of all pump,
find it; that your compound is really what you report it to be autosampler, and detector parameters directly from the
and not something else. Every spectral detail of your sample controller keypad, the PowerLine™ software embedded
is rapidly, easily, and accurately detected. Every peak is in the 600 controller allows you to store up to 15
analysed with mathematical precision so you can be sure to completed instrument methods for easy method retrieval.
asses the purity of peaks and confirm their identity. And you • By linking your 600E to the Millenium 2010
will never have to trade off resolution against sensitivity. Chromatography Manager, you can take advantage of
• The 996 PDA Detector lets you look at your results in a complete system control and information management.
variety of ways - even while a separation is under way.
• Absorbency measurement over the entire 190-800 nm
wavelength range.
• The MaxPlot feature of the Millenium PDA Software
automatically detects UV/Vis absorbing compounds
anywhere within the 190-800 nm range. It measures
their absorbence maxima, and plots a peak for each
compound at its absorbence maximum, creating a
composite chromatogram of all of the compounds it has
detected.
Waters™ 600E Multisolvent Delivery System
A no-compromise approach to single pump
gradient systems
Waters 600E Multisolvent Delivery System offers no-compro-
mise high-performance in a true single pump design. By
combining two Waters patented technologies, time-proven
solvent delivery and innovative Random Phase
Synchronization (RPS™) software, we can offer you
For more information
unmatched performance compared with any other single
pump gradient system. For more information on these
• An exceptionally wide total flow range from 0.01 to products check boxes 14-18
20.00 mL/min (with 45.0 mL/min optional) gives you on the Business Reply Card.
the versatility to perform analytical and semi-preparative
separations.
Carbonyl Compounds as DNPH Derivatives; EPA Method 554

Objective:
As more and more drinking water production
facilities turn to ozonation as a disinfectant, Column: Symmetry C18 5µm 3.9 mm x 150 mm
there has been a concomitant need for repro-
ducible and sensitive analytical methods for Mobile Phase: A: 50% acetonitrile in water
determination of oxonation related disinfection B: Acetonitrile
byproducts (DPBs) in finished waters. Among
Gradient: 100% A for 4 min,
the most important of these DPBs are aldehydes
Then linear gradient to 100% B in 25 min.
and ketones. EPA method 554 is a recom-
mended procedure for determination of aldehy- Flow Rate: 1.2 mL/min
des and ketones as dinitrophenylhydrazine Detection: UV @ 360 nm (0.12 AUFS)
derivatives by HPLC following solid phase
extraction. Injection Volume: 20 µL

Details:
1: Formaldehyde-DNPH 7: Pentanal-DNPH
This chromatographic procedure has been 2: Acetaldehyde-DNPH 8: Hexanal-DNPH
developed to provide a superior separation of
3: Propanal-DNPH 9:Heptanal-DNPH
the 12 EPA method 554 analytes in under 30
minutes. 4: Crotonaldehyde-DNPH 10: Octanal-DNPH

5: Butanal-DNPH 11: Nonanal-DNPH

6: Cyclohexanone-DNPH 12: Decanal-DNPH

System:

A Waters™ 600 S Controller, 616 pump, 712

WISP and 486 Detector were used for this
analysis. System Control, data acquisition and
reporting were accomplished using the
Millennium 2010 Chromatography Manager.

Waters Corporation 34 Maple Street, Milford, Massachusetts 01757-3696 U.S.A. 508 478-2000
Millennium 2010 Chromatography Manager The Waters 486 Tunable UV/VIS Absorbance
Detector
Advanced computer technology simplifies accessing,
interpreting, and reporting your results. The performance you expect with the flexibility you need.
The productivity of today’s chromatography is centered The Waters 486 Tunable UV/VIS Absorbance Detector is
around the ability to find, generate, summarise, and report the highest performing UV/VIS detector with the flexibility to
results in a way that meets your needs and complies with accommodate virtually all your HPLC applications. Our
any regulatory requirements. The Millenium relational data- exclusive Taper-Cell® flow cell design eliminates refractive
base lets you create your own view filters to customise the index effects, providing exceptional baseline stability.
way you sort your information, such as “lotCode”.
• The 486 detector is tunable from 190 to 600 nm
• Using the database, you can track a method history wavelength range with reproducibility and precision.
giving you audit trail capability that ensures total security
• The unique modular cell assemblies for microbore,
of your raw data and results. In addition, you can
analytical, preparative, high-pressure mass
specify “Access Type” which allows you to “lock” your
spectrometry, and nonmetallic flow paths give your
methods for even greater security.
application flexibility.
• The database keeps track of your instruments, triggering
an alarm when scheduled maintenance is required. The
Millenium relational database lets you view the injection
information on any calibration point at any particular
point in time on a processing sequence. This gives you
a detailed account of the processing method for that
injection, ensuring an accurate audit trail for regulatory
compliance.
• The Millenium Chromatography Manager can provide
acquisition and control for as many as four chromato-
graphic systems, including three HP5890 GC.
• The Millenium Report Publisher has the capability to give
you exactly what you want. For more information
For more information on these
products check boxes 14-18
on the Business Reply Card.
Determination of EPA Method 604 Phenols by HPLC

Objective:
Phenols are important industrial wastewater 9
contaminants which must be monitored for com- Column: Symmetry C8 5µm 3.9 mm x 150 mm
pliance with discharge regulations. Phenols, Mobile Phase: A: 1% acetic acid in water
particularly nitrophenols, often display inconsis- B: 1% acetic acid in acetonitrile
tent response when analysed by GC, even
5 Gradient: (T,%B)(0,30),(20,100)
when time consuming derivatisation steps are 4
employed. Flow Rate: 1.2 mL/min
Detection: UV @ 280 nm (0.08 AUFS)
Details:
2 Injection Volume: 40 µL
EPA Method 604 phenols are separated in
Concentration: 1.0 µg/mL each component
20 minutes using a Symmetry ® C8 5 µm
column. A Model 486 Variable Wavelength UV
Detector was used for this separation; for 6 1: Phenol 7: 4-Chloro-3-Methylphenol
3
greater sensitivity, a Model 460 pulsed 78 2: 4-Nitrophenol 8: 2,4-Dichlorophenol
electrochemical detector may by used. 1
10 3: 2-Chlorophenol 9: 4,6-Dinitro-2-Methylphenol

11 4: 2-Nitrophenol 10: 2,4,6-Trichlorophenol

5: 2,4-Dinitraphenol 11: Pentachlorophenol
6: 2,4-Dimethylphenol
0 5 10 15
Minutes

System:
A Waters™ 600 S Controller, 600 pump, 612
WISP Autosampler and a 486 detector was
used for this analyses. System control, data
acquisition and reporting were accomplished
using the Millennium 2010 Chromatography
Manager.

Waters Corporation 34 Maple Street, Milford, Massachusetts 01757-3696 U.S.A. 508 478-2000
Millennium 2010 Chromatography Manager The Waters 486 Tunable UV/VIS Absorbance
Detector
Advanced computer technology simplifies accessing,
interpreting, and reporting your results. The performance you expect with the flexibility you need.
The productivity of today’s chromatography is centered The Waters 486 Tunable UV/VIS Absorbance Detector is
around the ability to find, generate, summarise, and report the highest performing UV/VIS detector with the flexibility to
results in a way that meets your needs and complies with accommodate virtually all your HPLC applications. Our
any regulatory requirements. The Millenium relational data- exclusive Taper-Cell® flow cell design eliminates refractive
base lets you create your own view filters to customise the index effects, providing exceptional baseline stability.
way you sort your information, such as “lotCode”.
• The 486 detector is tunable from 190 to 600 nm
• Using the database, you can track a method history wavelength range with reproducibility and precision.
giving you audit trail capability that ensures total security
• The unique modular cell assemblies for microbore,
of your raw data and results. In addition, you can
analytical, preparative, high-pressure mass
specify “Access Type” which allows you to “lock” your
spectrometry, and nonmetallic flow paths give your
methods for even greater security.
application flexibility.
• The database keeps track of your instruments, triggering
an alarm when scheduled maintenance is required. The
Millenium relational database lets you view the injection
information on any calibration point at any particular
point in time on a processing sequence. This gives you
a detailed account of the processing method for that
injection, ensuring an accurate audit trail for regulatory
compliance.
• The Millenium Chromatography Manager can provide
acquisition and control for as many as four chromato-
graphic systems, including three HP5890 GC.
• The Millenium Report Publisher has the capability to give
you exactly what you want. For more information
For more information on these
products check boxes 14-18
on the Business Reply Card.