Optimization of a Mercury cold-vapor generation method

[Madsci]

Hi everyone,


I would like to optimize a Mercury Cold-Vapors analytical method which
isn't efficient to my opinion because we get broad and flat peaks.

Anyone would know what could cause theses very broad peaks whatever the
concentration I'm using for calibration (I usually use 0.1, 1, 5 and 10
PPB)? I suspect a kinetics problem.

AAS model: Buck 210 VPG with Graphite furnace and Cold-Vapor generation
options.

Argon flow : about 0,4 LPM

Reductant, SnCl2 10% w/v (10 grams of SnCl2 into 30 ml HCl, then I add about
1 gram of Mossy Tin, then I complete to 100 wih deionized water). Injected
into the reaction flask with a syringe and through a septum.

ACIDIFICATION SOLUTION: 20% v/v H2SO4 + 15% HCl diluted with deionized
water.

All our SAMPLES, STANDARDS AND BLANKS are being diluted with "preserved
water" : 0.2% HNO3 + 2ml / L Feldman solution (20%w/v K2CrO4 into 50%v/v
HNO3, in such a way we get a 0,04%w/v dichromate concentration in all
blanks, standards and samples)

For the reaction, we use 25ml of the sample, add 25 ml of the acidification
solution and add deionized water until volume reaches the 125 ml mark (to
make sure argon will bubble through the solution).

The moisture collector in-between the reaction flask and the quartz cell is
made of a plastic tube filled with 2-3 Kim-Wipes. No Magnesium perchlorate
is being used.


Anyone have an idea on what I should do to get nice and sharp peaks? Don,t
tell me to increase gas flow, I've already tried it!:)

Regards,


Daniel Picard, P.Chem.
Lab Manager
Aqua-Mac inc.
Montréal (Quebec)

[jose_gregorio_contreras via NatScience.com]

It is possible that the problem is in the amplification of the signal.

José

Post by Madsci
Hi everyone,
I would like to optimize a Mercury Cold-Vapors analytical method which
isn't efficient to my opinion because we get broad and flat peaks.
Anyone would know what could cause theses very broad peaks whatever the
concentration I'm using for calibration (I usually use 0.1, 1, 5 and 10
PPB)? I suspect a kinetics problem.
AAS model: Buck 210 VPG with Graphite furnace and Cold-Vapor generation
options.
Argon flow : about 0,4 LPM
Reductant, SnCl2 10% w/v (10 grams of SnCl2 into 30 ml HCl, then I add about
1 gram of Mossy Tin, then I complete to 100 wih deionized water). Injected
into the reaction flask with a syringe and through a septum.
ACIDIFICATION SOLUTION: 20% v/v H2SO4 + 15% HCl diluted with deionized
water.
All our SAMPLES, STANDARDS AND BLANKS are being diluted with "preserved
water" : 0.2% HNO3 + 2ml / L Feldman solution (20%w/v K2CrO4 into 50%v/v
HNO3, in such a way we get a 0,04%w/v dichromate concentration in all
blanks, standards and samples)
For the reaction, we use 25ml of the sample, add 25 ml of the acidification
solution and add deionized water until volume reaches the 125 ml mark (to
make sure argon will bubble through the solution).
The moisture collector in-between the reaction flask and the quartz cell is
made of a plastic tube filled with 2-3 Kim-Wipes. No Magnesium perchlorate
is being used.
Anyone have an idea on what I should do to get nice and sharp peaks? Don,t
tell me to increase gas flow, I've already tried it!:)
Regards,
Daniel Picard, P.Chem.
Lab Manager
Aqua-Mac inc.
Montréal (Quebec)

[d***@chem.utoronto.ca]

Post by Madsci
Hi everyone,
I would like to optimize a Mercury Cold-Vapors analytical method which
isn't efficient to my opinion because we get broad and flat peaks.
Anyone would know what could cause theses very broad peaks whatever the
concentration I'm using for calibration (I usually use 0.1, 1, 5 and 10
PPB)? I suspect a kinetics problem.
AAS model: Buck 210 VPG with Graphite furnace and Cold-Vapor generation
options.
Argon flow : about 0,4 LPM
Reductant, SnCl2 10% w/v (10 grams of SnCl2 into 30 ml HCl, then I add about
1 gram of Mossy Tin, then I complete to 100 wih deionized water). Injected
into the reaction flask with a syringe and through a septum.
ACIDIFICATION SOLUTION: 20% v/v H2SO4 + 15% HCl diluted with deionized
water.
All our SAMPLES, STANDARDS AND BLANKS are being diluted with "preserved
water" : 0.2% HNO3 + 2ml / L Feldman solution (20%w/v K2CrO4 into 50%v/v
HNO3, in such a way we get a 0,04%w/v dichromate concentration in all
blanks, standards and samples)
For the reaction, we use 25ml of the sample, add 25 ml of the acidification
solution and add deionized water until volume reaches the 125 ml mark (to
make sure argon will bubble through the solution).
The moisture collector in-between the reaction flask and the quartz cell is
made of a plastic tube filled with 2-3 Kim-Wipes. No Magnesium perchlorate
is being used.
Anyone have an idea on what I should do to get nice and sharp peaks? Don,t
tell me to increase gas flow, I've already tried it!:)
Regards,
Daniel Picard, P.Chem.
Lab Manager
Aqua-Mac inc.
Montréal (Quebec)

Try a heating mantle. The quartz cell is placed in the heating mantle
and the temperature along the length of the cell is maintained at
about 10 oC above ambient. This will prevent condensation within the
quartz cell. It may sharpen up your peaks

DTM

[Madsci]

Hi Dan,

Interesting! Actually, I read a paper where someone was using a desk lamp
with an incadescent light bulb (not theses fluorescent lamps, or you would
suddenly see an increase in your signal :) ) to heat up the quartz cell,
thus reducing water condensation into it. It may help, but I haven't seen
any water condensation into the cell yet.

However, as for many things in chemistry... The most important is unseen to
our poor eyes :)

Thank you for the idea,


Daniel Picard, P.Chem.

P.S. I really get very broad peaks. Increasing voltage gain on my phototube
have helped, but I know it can still be improved. And, honestly, theses Buck
Scientific AAS are very.... very..... frustrating!

Post by Madsci
Hi everyone,
I would like to optimize a Mercury Cold-Vapors analytical method which
isn't efficient to my opinion because we get broad and flat peaks.
Anyone would know what could cause theses very broad peaks whatever the
concentration I'm using for calibration (I usually use 0.1, 1, 5 and 10
PPB)? I suspect a kinetics problem.
AAS model: Buck 210 VPG with Graphite furnace and Cold-Vapor generation
options.
Argon flow : about 0,4 LPM
Reductant, SnCl2 10% w/v (10 grams of SnCl2 into 30 ml HCl, then I add about
1 gram of Mossy Tin, then I complete to 100 wih deionized water). Injected
into the reaction flask with a syringe and through a septum.
ACIDIFICATION SOLUTION: 20% v/v H2SO4 + 15% HCl diluted with deionized
water.
All our SAMPLES, STANDARDS AND BLANKS are being diluted with "preserved
water" : 0.2% HNO3 + 2ml / L Feldman solution (20%w/v K2CrO4 into 50%v/v
HNO3, in such a way we get a 0,04%w/v dichromate concentration in all
blanks, standards and samples)
For the reaction, we use 25ml of the sample, add 25 ml of the
acidification
solution and add deionized water until volume reaches the 125 ml mark (to
make sure argon will bubble through the solution).
The moisture collector in-between the reaction flask and the quartz cell is
made of a plastic tube filled with 2-3 Kim-Wipes. No Magnesium perchlorate
is being used.
Anyone have an idea on what I should do to get nice and sharp peaks? Don,t
tell me to increase gas flow, I've already tried it!:)
Regards,
Daniel Picard, P.Chem.
Lab Manager
Aqua-Mac inc.
Montréal (Quebec)

Try a heating mantle. The quartz cell is placed in the heating mantle
and the temperature along the length of the cell is maintained at
about 10 oC above ambient. This will prevent condensation within the
quartz cell. It may sharpen up your peaks

DTM

[Samite Alchemist]

Are you sure you are supposed to get peaks? Generally this technique
gives a steady state reading, like AA or ICP, not a peak.

Samite Alchemist

[Shankar Bhattacharyya]

Post by Samite Alchemist
Are you sure you are supposed to get peaks? Generally this
technique gives a steady state reading, like AA or ICP, not a peak.

I am aware of two, maybe three basic designs for the equipment,
depending on how you count. My count is three.

You have closed systems, "powered" by an in-line air pump. The pump
bubbles air through a reaction flask and circulates it through the
closed loop, which includes the optical cell. The signal rises to a
shortish equilibrium concentration in the headspace of the system,
plateaus, then declines at a rate dependent on the lekage from the
system and various reaction rate related variables. We used to do the
work with such a system, home-made with Kontes glassware, an
overpriced aquarium pump and tygon tubing. We used to have lots of
trouble with that, mostly related to the leaks inherent in such a
system. Ground-glass joints of the modern quickfit/multifit type
simply do not fit as well as they used to. The benefit, of course, is
that they fit much better in arbitrary pairs than the old ones used
to. The loss is that the old-ground-as-a-mating-pair joints did fit
very well as a fixed pair.

You have open-tube systems, where the reactants are in a vessel of
some kind, and a pulse of airflow takes a "peak" through the tube.

You have an open tube system where sample and reagents are introduced
as a constant stream and the mercury so generated is blown into an
open optical tube. We use such a system now, a Varian VG-7. It uses a
peristaltic pump with different id tubing for the various reagents.
It uses a lot of sample extract but it produces stunningly flat
response. Ruler-flat for a good many seconds. We have zero problems
with it. It is also five or ten times as sensitive as the closed
system we used before.

- Shankar