b. Column Equilibration
c. Initial Column Efficiency Determination
II. COLUMN USE
a. Guard Columns
b. Sample Preparation
c. Operating pH Limits
d. Solvents
e. Pressure
f. Temperature
III. SCALING UP/DOWN ISOCRATIC METHODS
IV. TROUBLESHOOTING
V. COLUMN CLEANING, REGENERATION
AND STORAGE
a. Cleaning and Regeneration
b. Storage
Thank you for choosing a Waters XSelect® CSH column. This HPLC
column features Waters Charged Surface Hybrid (CSH) Technology
which provides excellent peak shape, high efficiency and loading
capacity for basic compounds when using low ionic strength, acidic
mobile phases. T his same particle technology is used in the ACQUITY
UPLC
between HPLC and UPLC® system platforms. All XSelect CSH packing
materials are manufactured in a cGMP, ISO 9001:2000 certified plant
using ultra pure reagents. Each batch of XSelect CSH material is
tested chromatographically with acidic, basic and neutral analytes
and the results are held to narrow specification ranges to assure
excellent, reproducible performance. Every column is individually
tested and a Performance Test Chromatogram is provided with
each column along with the Certificate of Acceptance.
VI. CONNECTING THE COLUMN TO THE HPLC
a. Column Connectors and System Tubing Considerations
b. Measuring System Band-spreading Volume and System Variance
c. Measuring Gradient Delay Volume (or Dwell Volume)
VII. ADDITIONAL INFORMATION
a. Use of Narrow-Bore (3.0 mm i.d.) Columns
b. Impact of Band-spreading Volume on 2.1 mm i.d.
Column Performance
c. Non-Optimized vs. Optimized LC/MS/MS System:
System Modification Recommendations
Page 2
[ CARE AND USE MANUAL ]
I. GETTING STARTED
Each XSelect CSH column comes with a Certificate of Analysis and
a Performance Test Chromatogram. The Certificate of Analysis is
specific to each batch of packing material contained in the XSelect
CSH column and includes the batch number, analysis of unbonded
particles, analysis of bonded particles, and chromatographic
results and conditions. The Performance Test Chromatogram is
specific to each individual column and contains information such
as: batch number, column serial number, USP plate count, USP
tailing factor, retention factor, and chromatographic conditions.
This data should be stored for future reference.
a. Column Installation
Note: The flow rates given in the procedure below are for a typical 5 µm packing
in a 4.6 mm i.d. column. Scale the flow rate up or down accordingly based upon
the i.d., length, particle size and backpressure of the XSelect CSH column being
installed. See Scaling Up/Down Isocratic Separations section for calculating
flow rates when changing column i.d and/or length. See Connecting the Column
to the HPLC section for a more detailed discussion on HPLC connections.
1. Purge the pumping system of any buffer-containing mobile phases
and connect the inlet end of the column to the injector outlet. An
arrow on the column identification label indicates the correct
direction of solvent flow.
b. Column Equilibration
XSelect CSH columns are shipped in 100% acetonitrile. It is
important to ensure mobile-phase compatibility before changing
to a different mobile-phase system. Equilibrate the column with
a minimum of 10 column volumes of the mobile phase to be used
(refer to Table 1 for a listing of empty column volumes).
To avoid precipitating out mobile-phase buffers on your column
or in your system, flush the column with five column volumes of
a water/organic solvent mixture, using the same or lower solvent
content as in the desired buffered mobile phase. (For example,
flush the column and HPLC system with 60% methanol in water
prior to introducing 60% methanol/40% buffer mobile phase).
c. Initial Column Efficiency Determination
1. Perform an efficiency test on the column before using it in the desired
application. Waters recommends using a suitable solute mixture,
as found in the “Performance Test Chromatogram”, to analyze the
column upon receipt.
2. Determine the number of theoretical plates (N) and use t his value for
periodic comparisons.
2. Flush column with 100% organic mobile phase (methanol or
acetonitrile) by setting the pump flow rate to 0.1 mL/min and
increase the flow rate to 1 mL/min over 5 minutes.
3. When the mobile phase is flowing freely from the column
outlet, stop the flow and attach the column outlet to the
detector. This prevents entry of air into the detection system
and gives more rapid baseline equilibration.
4. Gradually increase the flow rate as described in step 2.
5. Once a steady backpressure and baseline have been achieved,
proceed to the next section.
Note: If mobile-phase additives are present in low concentrations (e.g., ion-pairing
reagents), 100 to 200 column volumes may be required for complete equilibration. In
addition, mobile phases that contain formate (e.g., ammonium formate, formic acid,
etc.) may also require longer initial column equilibration times.
3. Repeat the test at predetermined intervals to track column
performance over time. Slight variations may be obtained on
two different HP LC systems due to the quality of the connections,
operating environment, system electronics, reagent quality, column
condition and operator technique.
Table 1: Empty Column Volumes in mL (multiply by 10 for flush
solvent volumes)
To ensure the continued high performance of XSelect CSH columns, follow these guidelines:
a. Guard Columns
Use a Waters guard column of matching chemistry and particle size between the injector and main column. It is important to use a highperformance matching guard column to protect the main column while not compromising or changing the analytical resolution.
Guard columns need to be replaced at regular intervals as determined by sample contamination. When system backpressure steadily
increases above a set pressure limit, it is usually an indication that the guard column should be replaced. A sudden appearance of split
peaks is also indicative of a need to replace the guard column.
b. Sample Preparation
1. Sample impurities often contribute to column contamination. One option to avoid this is to use Waters Oasis® solid-phase extraction cartridges/columns
or Sep-Pak® cartridges of the appropriate chemistry to clean up the sample before analysis.
2. It is preferable to prepare t he sample in the operating mobile p hase or a mobile phase that is weaker (less organic modifier) than the mobile phase for the
best peak shape and sensitivity.
3. If the sample is not dissolved in the mobile phase, ensure that the sample, solvent and mobile phases are miscible in order to avoid sample and/or buffer
precipitation.
4. Filter sample with 0.2 µm filters to remove particulates. If the sample is dissolved in a solvent that contains an organic modifier (e.g.,
acetonitrile, methanol, etc.) ensure that the filter material does not dissolve in the solvent. Contact the filter manufacturer with solvent
compatibility questions. Alternatively, centrifugation for 20 minutes at 8,000 rpm, followed by the transfer of the supernatant liquid to an
appropriate vial, could be considered.
c. Operating pH Limits
The recommended operating pH limits for XSelect CSH columns are listed in Table 2. A listing of commonly used buffers and additives is given
in Table 3. Additionally, the column lifetime will vary depending upon the operating temperature, type and concentration of buffer used.
Table 2: Recommended pH and Temperature Limits for XSelect CSH columns
Columnn NameParticle Size (µm)Pore Diameter (Å)Surface Area (m2)pH Limits
Ion pair additive, can suppress MS signal, used in the 0.02-0.1% range.
Maximum buffering obtained when used with ammonium acetate salt. Used in 0.1-1.0% range.
Maximum buffering obtained when used with ammonium formate salt. Used in 0.1-1.0% range.
Used in the 1-10 mM range. Note that sodium or potassium salts are not volatile.
Used in the 1-10 mM range. Note that sodium or potassium salts are not volatile.
Traditional low pH buffer, good UV transparency.
Above pH 7, reduce temperature/concentration and use a guard column to maximize lifetime.
Above pH 7, reduce temperature/concentration and use a guard column to maximize lifetime.
Generally used at 10 mM or less.
Used in the 5-10 mM range (for MS work keep source >150 ˚C ). Adjust pH with ammonium
hydroxide or acetic acid. Good buffering capacity at pH 10.
Note: use ammonium bicarbonate (NH4HCO3), not ammonium carbonate ((NH4)2CO3)
Used in the 1-10 mM range.
Used in the 1-10 mM range.
Reduce temperature/concentration and use a guard column to maximize lifetime.
Zwitterionic buffer, compatible with acetonitrile, used in the 1-10 mM range. Low odor.
Zwitterionic buffer, can give longer lifetimes than borate buffer.
Used in the 1-10 mM range.
Zwitterionic buffer, compatible with acetonitrile, used in the 1-10 mM range. Low odor.
Used in the 0.1-1.0% range. Volatile only when titrated with acetic acid (not hydrochloric or
phosphoric). Used as ion-pair for DNA analysis at pH 7-9.
Mild buffer, gives long lifetime.
d. Solvents
To maintain maximum column performance, use high quality chromatography grade solvents. Filter all aqueous buffers prior to use. Pall
Gelman Laboratory Acrodisc® filters are recommended. Solvents containing suspended particulate materials will generally clog the outside
surface of the inlet distribution frit of the column. This will result in higher operating pressure and poor performance.
Degas all solvents thoroughly before use to prevent bubble formation in the pump and detector. The use of an on-line degassing unit is also
recommended. This is especially important when running low pressure gradients since bubble formation can occur as a result of aqueous
and organic solvent mixing during the gradient.
e. Pressure
XSelect CSH columns can tolerate pressures of up to 6,000 psi (400 bar or 40 Mpa) although pressures greater than 4,000 – 5,000 psi
should be avoided in order to maximize column and system lifetimes.
f. Temperature
Temperatures up to 80 ˚C are recommended for operating XSelect CSH columns in order to enhance selectivity, lower solvent viscosity and
increase mass transfer rates. However, any temperature above ambient will have a negative effect on lifetime which will vary depending on
the pH and buffer conditions used. See Table 2 for recommended pH and temperature operating ranges.
4
XSelect CSH HPLC Columns
Page 5
[ CARE AND USE MANUAL ]
III. SCALING UP/DOWN ISOCRATIC METHODS
The following formulas will allow scale up or scale down, while
maintaining the same linear velocity, and provide new sample
loading values:
If column i.d. and length are altered:
F2 = F1 (r2/r1)2
Load2 = Load1 (r2/r1)2(L2/L1)
Injection volume2 = Injection volume1(r2/r1)2(L2/L1)
Where: r = Radius of the column
F = Flow rate
L = Length of column
1 = Original, or reference column
2 = New column
IV. TROUBLESHOOTING
Changes in retention time, resolution, or backpressure are
often due to column contamination. See the Column Cleaning,
Regeneration and Storage section of this Care and Use Manual.
Information on column troubleshooting problems may be found
in HPLC Columns Theory, Technology and Practice, U.D. Neue,
(Wiley-VCH, 1997), the Waters HPLC Troubleshooting Guide
(Literature code # 720000181EN) or visit the Waters Corporation
website for information on seminars (www.waters.com).
V. COLUMN CLEANING, REGENERATION,
AND STORAGE
a. Cleaning and Regeneration
Changes in peak shape, peak splitting, shoulders on the
peak, shifts in retention, change in resolution or increasing
backpressure may indicate contamination of the column. Flushing
with a neat organic solvent, taking care not to precipitate buffers,
is usually sufficient to remove the contaminant. If the flushing
procedure does not solve the problem, purge the column using the
following cleaning and regeneration procedures.
Use the cleaning routine that matches the properties of the
samples and/or what you believe is contaminating the column
(see Table 4). Flush columns with 20 column volumes each of
HPLC-grade solvents (e.g., 80 mL total for 4.6 x 250 mm column)
listed in Table 4. Increasing mobile phase temperature to 35-55
˚C increases cleaning efficiency. If the column performance is poor
after cleaning and regeneration, call your local Waters office for
additional support.
Table 4: Cleaning and Regeneration Sequence or Options
*Use low organic solvent content to avoid precipitating buffers.
(or an appropriate
isopropanol/water mixture*)
3. Dichloromethane
5. Isopropanol
(followed by an appropriate
isopropanol/water mixture*)
Option 1: Inject repeated
aliquots of dimethyl sulfoxide
(DMSO)
Option 2: gradient of 10% to
90% B where:
A = 0.1% trifluoroacetic acid
(TFA) in water
B = 0.1% trifluoroacetic acid
(TFA) in acetonitrile (CH
Option 3: Flush column with 7M
guanidine hydrochloride, or
7M urea
CN)
3
b. Storage
For periods longer than four days at room temperature, store
XSelect CSH columns in 100% acetonitrile. Immediately after
use with elevated temperatures and/or at pH extremes, store
in 100% acetonitrile for the best column lifetime. Do not store
columns in highly aqueous (<20% organic) mobile phases, as this
may promote bacterial growth. If the mobile phase contained a
buffer salt, flush the column with 10 column volumes of HPLC
grade water (see Table 1 for common column volumes) and
replace with 100% acetonitrile for storage. Failure to perform this
intermediate step could result in precipitation of the buffer salt
in the column or system when 100% acetonitrile is introduced.
Completely seal column to avoid evaporation and drying out of the
packed bed.
5
XSelect CSH HPLC Columns
Page 6
[ CARE AND USE MANUAL ]
Void
VI. CONNECTING THE COLUMN TO THE HPLC
a. Column Connectors and System Tubing Considerations
Tools needed:
• 3/8 inch wrench
• 5/16 inch wrench
Handle the column with care. Do not drop or hit the column on a
hard surface as it may disturb the bed and affect its performance.
1. Correct connection of 1/16 inch outer diameter stainless steel
tubing leading to and from the column is essential for high-quality
chromatographic results.
2. When using standard stainless steel compression screw fittings,
it is important to ensure proper fit of the 1/16 inch outer diameter
stainless steel tubing. W hen tightening or loosening the compression
screw, place a 5/16 inch wrenc h on the compression screw and a 3/8
inch wrench on the hex head of the column endfitting.
Note: If one of the wrenches is placed on the column tube flat during this
process, the endfitting will be loosened and leak.
3. If a leak occurs between the stainless steel compression screw fitting
and the column endfitting, a new compression screw fitting, tubing
and ferrule must be assembled.
endfittings (Figure 1). Each endfitting style varies in the required
length of the tubing protruding from the ferrule. The XSelect CSH
column is equipped with Waters style endfittings that require a
0.130 inch ferrule depth. If a non-Waters style column is presently
being used, it is critical that ferrule depth be reset for optimal
performance prior to installing an XSelect CSH column.
In a proper tubing/column connection (Figure 2), the tubing touches
the bottom of the column endfitting, with no void between them.
Figure 2: Proper Tubing/Column Connection
The presence of a void in the flow stream reduces column
performance. This can occur if a Parker ferrule is connected to a
Waters style endfitting (Figure 3).
Note: A void appears if tubing with a Parker ferrule is connected to a Waters
style column.
4. An arrow on the column identification label indicates correct direction
of solvent flow.
Correct connection of 1/16 inch outer diameter stainless steel
tubing leading to and from the column is essential for high-quality
chromatographic results. To obtain a void-free connection, the tubing
must touch the bottom of the column endfitting. It is important to
realize that extra column peak broadening due to voids can destroy
an otherwise successful separation. The choice of appropriate column
connectors and system tubing is discussed in detail below.
Waters Ferrule SettingParker Ferrule Setting
.130”.090”
Figure 1: Waters and Parker Ferrule Types
Due to the absence of an industry standard, various column
manufacturers have employed different types of chromatographic
column connectors. The chromatographic separation can be
negatively affected if the style of the column endfittings does not
match the existing tubing ferrule settings. This section explains
the differences between Waters style and Parker style ferrules and
Figure 3: Parker Ferrule in a Waters Style Endfitting
There is only one way to fix this problem: Cut the end of the tubing
with the ferrule, place a new ferrule on the tubing and make a new
connection. Before tightening the screw, make sure that the tubing
bottoms out in the endfitting of the column.
Conversely, if tubing with a Waters ferrule is connected to a
column with Parker style endfitting, the end of the tubing will
bottom out before the ferrule reaches its proper sealing position.
This will leave a gap and create a leak (Figure 4).
Note: The connection leaks if a Water ferrule is connected to a column with a
Parker style endfitting.
Gap
Figure 4: Waters Ferrule in a Parker Style Endfitting
6
XSelect CSH HPLC Columns
Page 7
[ CARE AND USE MANUAL ]
There are two ways to fix the problem:
1. Tighten the screw a bit more. The ferrule moves forward, and reaches
the sealing surface. Do not over tighten since this may end in breaking
the screw.
2. Cut the tubing, replace the ferrule and make a new connection.
Alternatively, replace the conventional compression screw fitting
with an all-in-one PEEK fitting (Waters part number PSL613315)
that allows resetting of the ferrule depth. Another approach is to
use a SLIPFREE® connector to always ensure the correct fit. The
fingertight SLIPFREE® connectors automatically adjust to fit all
compression screw type fittings without the use of tools (Figure 5).
Table 5: Waters Part Numbers for SLIPFREE® Connectors
Figure 6 shows the influence of tubing internal diameter on system
band spreading and peak shape. As can be seen, the larger tubing
diameter causes excessive peak broadening and lower sensitivity.
0.005 inches
0.020 inches
0.040 inches
Figure 5: Single and Double SLIPFREE® Connectors
SLIPFREE® Connector Features:
• Tubing pushed into endfitting, thereby guaranteeing a void-free
connection
• Connector(s) come(s) installed on tubing
• Various tubing IDs and lengths available
• Fingertight to 10,000 psi – never needs wrenches
• Readjusts to all column endfittings
• Compatible with all commercially available endfittings
• Unique design separates tube- holding function from sealing function
Diluted/Distorted Sample Band
Figure 6: Effect of Connecting Tubing on System
b. Measuring System Band-spreading Volume and
System Variance
This test should be performed on an HPLC system with a single
wavelength UV detector (not a Photodiode Array (PDA)).
1. Disconnect column from system and replace with a zero dead
volume union.
2. Set flow rate to 1 mL/min.
3. Dilute a test mix in mobile phase to give a detector sensitivity of
0.5 - 1.0 AUFS (system start up test mix can be used which contains
uracil, ethyl and propyl parabens; Waters part number WAT034544).
4. Inject 2 to 5 µL of this solution.
7
XSelect CSH HPLC Columns
Page 8
[ CARE AND USE MANUAL ]
5. Measure the peak width at 4.4% of peak height (5-sigma method):
5-sigma Band-spreading (µL) = Peak Width (min) x Flow Rate (mL /min) x
(1000 µL/1 mL)
System Variance (µL
System Volume
2
) = (5-sigma band-s preading)2/ 25
5
4.4 %h
Figure 7: Determination of System Band-spreading Volume Using 5-Sigma Method
In a typical HPLC system, the Band-spreading Volume should be
no greater than 100 µL ± 30 µL (or Variance of 400 µL2 ± 36 µL2).
In a microbore (2.1 mm i.d.) system, the Band-spreading Volume
should be no greater than 20 to 40 µL (or Variance no greater
than 16 µL2 to 64 µL2).
c. Measuring Gradient Delay Volume (or Dwell Volume)
For successful gradient-method transfers the gradient delay
volumes should be measured using the same method on both HPLC
systems. The procedure below describes a method for determining
the gradient delay volumes.
1. Replace the column with a zero dead volume union.
2. Prepare mobile phase A (pure solvent, such as methanol) and
mobile phase B (mobile phase A with a UV absorbing sample,
such as (v/v) 0.1% acetone in methanol).
3. Equilibrate the system with mobile phase A until a stable
baseline is achieved.
4. Set the detector wavelength to the absorbance maximum of the
probe (265 nm for acetone).
5. Program a 0-100% B linear gradient in 10 min at 2 mL/
min (the exact conditions are not critical; just make sure the
gradient volume is at least 20 mL) with a hold at 100% B.
Figure 8: Determination of Gradient Delay Volume
6. Determine the dwell time by first loc ating the time at the midpoint of
the formed gradient (t
) (half the vertical distance between the initial
1/2
and final isocratic segments as shown in Figure 8).
7. Subtract half the gradient time (1/2 tg) (10 min/2 = 5 min in this
example) from the gradient midpoint (t
) to obtain the dwell time (tD).
1/2
8. Convert the dwell time (tD) to the dwell volume (VD) by multiplying by
the flow rate (F).
Dwell Volume VD = (t
1/2
- 1/ 2
) x F
tg
For fast gradient methods, the gradient delay volume (or dwell
volume) should be less than 1 mL. If the gradient delay volume
is greater than 1 mL, see System Modification Recommendations
section on how to reduce system volume.
VII. ADDITIONAL INFORMATION
a. Use of Narrow-Bore (<3.0 mm i.d.) Columns
This section describes how to minimize extra column effects
and provides guidelines on maximizing the performance of a
narrow-bore column in an HPLC system. A 3.0 mm i.d. narrowbore column usually requires no system modifications. A 2.1 mm
i.d. column, however, requires modifications to the HPLC system
in order to eliminate excessive system band-spreading volume.
Without proper system modifications, excessive system bandspreading volume causes peak broadening and has a large impact
on peak width as peak volume decreases.
8
XSelect CSH HPLC Columns
Page 9
[ CARE AND USE MANUAL ]
b. Impact of Band-spreading Volume on 2.1 mm i.d.
Column Performance
System with 70 µL band-spreading: 10,000 plates
System with 130 µL band-spreading: 8,000 plates
(same column)
Note: Flow splitters after the column will introduce additional band-spreading.
System optimization, especially in a system that contains a flow
splitter, can have dramatic effects on sensitivity and resolution.
Optimization includes using correct ferrule depths and minimizing
tubing inner diameters and lengths. An example is given in Figure
9 where system optimization resulted in a doubling of sensitivity
and resolution of the metabolite in an LC/MS/MS system.
c. Non-Optimized vs. Optimized LC/MS/MS System:
System Modification Recommendations
1. Use a microbore detector flow cell with 2.1 mm i.d. columns.
Note: Detector sensitivity is reduced with the shorter flow cell path length
in order to achieve lower band-spreading volume.
2. Minimize injector sample loop volume.
3. Use 0.009 inch (0.25 mm) tubing for rest of connections
in standard systems and 0.005 inch (0.12 mm) tubing for
narrowbore (2.1 mm i.d.) systems.
4. Use perfect (pre-cut) connections (with a variable depth inlet
if using columns from different suppliers).
5. Detector time constants should be shortened to less than
0.2 seconds.
7.007.50
Non-optimized LC/MS/MS SystemOptimized System
Figure 9: Non-Optimized vs. Optimized LC/MS/MS System