icp system configuration and considerations for plant
TRANSCRIPT
www.spectro.com
ICP system configuration and
considerations for plant foods and
fertilizers
Dion Tsourides
Spectro A.I. Inc
91 McKee Dr
Mahwah, NJ
www.spectro.com
Best plasma options for fertilizers or agronomic samples???
•Radial plasma viewing offers distinct
advantages for these matrices.
•Dual View (DV) systems promise
“the best of both worlds” but
analytically fall short.
•IF axial viewing may be needed,
there is a far better option than DV.
•ICP MS = Research grade tool that
has issues with % level solutions BUT
is an excellent option for research
and trace elements (heavy metals).
It’s a complimentary technique.
•Radial Plasma:
1.Better linearity, stability and
precision in heavy matrices.
2.Higher sample throughput (less
rinse, shorter integration times,
less matrix concerns).
3.Lower cost of ownership
(torches, less internal stds – Cs
not required, a saving of up to
$6k/year)
4.Reduced “mechanical” error
(less dilution, less matrix
modifications (Cs/Li addition), no
cone/interface maintenance, etc.)
www.spectro.com
Experimental
• Radial ICP OES
• Cyclonic Chamber with Seapray nebulizer, 1.8mm injector
• 1450 W, 14 L/min Coolant, 1.2 L/min Aux, 0.8 L/min Nebulizer, (+
0.2 L/min Aux)
• Standard Calibration
• Empirical Calibration
• Varying Internal Standards
• Various Calibration standards
• Digestion concerns (P volatility)
www.spectro.com
Optional Gas adapter (laminar flow to force sample into
plasma and avoid external cooling of plasma).
www.spectro.com
Anatomy of an active plasma stream
• High Na matrix to show the
plasma has a “skin depth
effect” which causes much of
the sample to pass around the
plasma and not complete
through the plasma.
• Additional gas option forces
the sample stream through the
plasma.
www.spectro.com
Accuracy controls
1. Linearity : An ICP (all ICP's) typically have +/- 2% error in the
calibration curve. What does that mean? Using the exact
same conditions, intro system and solutions...each time you
calibrate, the accuracy of the curve can vary up to 2%. So, on
a 50% check solution a range of 49-51% is expected.
2. Adding empirical calibration controls can reduce this to 0.5%
or less.
3. Additional gas flow (intro system) can produce a more
consistent sample delivery with less plasma effects
(efficiency/cooling/EIE).
4. Internal Standard selection and error from using a Blank.
5. Wavelength averaging.
www.spectro.com
Method defined calibration with Additional Gas
Sample P2O5 Expected
% %
10B ! 0.012 0
7B 51.62 51.34
5B 46.029 46.32
3B 40.54 40.94
www.spectro.com
Comments and recommendations
1. Additional
carrier gas
seems to help
accuracy.
2. In a traditional
calibration the
biggest limiting
factor is the
ICP's linear
error (2%).
3. Empirical
calibrations help
accuracy and long
term precision....!
Wt./Wt. Dilution
0.5012g/0.5g
6. Varying tubing
sizes and dilution
rate helps based
on specific
instrumentation
more than on the
methodology.
5. Internal
standard,
wavelength
selection, and
averaging also
help accuracy.
4. Calibration
standards
(species) affect
accuracy as
does digestion!
www.spectro.com
Axial Plasma Observation – Some historical perspective
• Axial plasma observation was introduced in the mid 1990s
• The development driven by requirements for higher sensitivity,
particularly for the toxic, heavy metals
• Axial plasma observation still plays an important role as it provides
much higher matrix tolerance as compared to ICP-MS for majors (P
& K)
• ICP MS provides better trace elemental analysis, primarily for the
heavy metals.
13
www.spectro.com
Axial Plasma Observation – Challenges
• Axial plasma observation improves the sensitivity by sampling the
emitted light from the entire excitation channel
• However, it also suffers from stronger influences, since all
phenomena present in the excitation are viewed
• Pros and Cons: – High sensitivity
– Stronger matrix effects
– Less suitable for high TDS and
applications with organic solutions
14
Viewing volume
www.spectro.com
Radial Plasma Observation – The choice for high stability, high matrix loads and organic solutions
• Radial plasma observation provides lower sensitivity since the
central channel is only partially view...or does it? More on this later!
• However, it provides high stability and freedom from matrix effects
since the affected zones in the plasma are blanked out
• Pros and Cons: – High stability
– High matrix tolerance & LOWER Maintenance
– High linear dynamic range
– Freedom from matrix effects
– Lower sensitivity.....???
– It depends on the optics and interface used!
Viewing volume
www.spectro.com
“Dual View” – perception vs reality
• “Dual View” appears to solve the dilemma having to make a choice
between axial and radial plasma observation
•But!
• Only one view uses the direct light path, the other is
compromised...REGARDLESS OF HOW THE TORCH IS
MOUNTED!!!
Sensitivity loss due to additional optical
components
Sensitivity loss in the UV since the
light path cannot be purged effectively
Higher Straylight, Higher maintenance
16
Coupling mirror Optic window
Mirror, plain Mirror, concave
Torch
Entrance slit optic
www.spectro.com 17
TRUE RADIAL SYSTEM versus a DUAL VIEW SYSTEM in a VERTICAL
CONFIGURATION, is sensitivity effected?
TRUE RADIAL (no
mirrors, periscope, no
interfaces/cones)
Dual View, Twin Interface, DSDV, etc.
www.spectro.com
“Dual View” – continued
• For highest sensitivity, instrument using a “Dual View” technique
typically have a horizontal torch orientation
Disadvantages with higher matrix loads and organic solutions
• Compared to a dedicated radial interface the sampling volume
using a periscope is smaller
A higher level of stability and precision as compared to a dedicated
radial systems cannot be achieved
The “dual view” technique serves a purpose, but does not provide
the highest possible performance in feeds and fertilizers
If using a DV system, the RADIAL mode performance should be fully
vetted BEFORE moving any elements (especially Grp I and II) to the
axial mode!
18
www.spectro.com 20
So when optimizing and selecting a system for
fertilizers, one should always consider the
requirements of the analytical task OVER the
“perceived” advantages of the hardware because in
optical emission spectroscopy, things are not always
as they seem.
Thank you for your time!