Problems using an HP 1050 pump with glycerol as mobile phase

[davtuner]

Hello,

First, let me say the title may be a bit misleading. I am not doing HPLC, but I am trying to use an HP1050 HPLC pump to pump glycerol into a reaction system. The pump works fine using water, methanol, etc but is having some problems pumping glycerol. I am trying to flow at 0.333 ml/minute into a system at vacuum. At first, I tried heating the glycerol in the solvent bottle to help lower the viscosity, to no avail. The high surface area/volume ratio of the smaller diameter tubings caused the sample to cool down before it even reached the piston block.

Then, I thought the problem may have been that the tubing was unable to deliver glycerol to the first piston, even if the first piston was pulling full vacuum pressure in the piston chamber. I addressed this by pressurizing (in a rated pressure vessel) the headspace of the glycerol feed bottle to 160 psi. This allowed the pump to flow some glycerol, but eventually the pump maxed out on pressure.

To address the maxing out on outlet pressure, I removed the glass filter frit on the outlet valve, and it brought the pressure down to a reasonable level.

Then, I noticed the pumping rate was not linear with respect to indicated pumping rate. For instance, at 0.5 ml/min the pump put out a droplet of glycerol every 5.5 seconds. At 0.25 ml/min the pump put out a droplet every 7.5 seconds. The droplets have roughly the same volume, and are a good estimate of volumetric flow.

Tl;dr : Does anyone have any suggestions on how to use this pump with a very viscous solution such as glycerol, and get a good correspondance between indicated flow/actual flow? Thanks.

[Consumer Products Guy]

The 1050 and other HPLC pumps are NOT designed for liquids of this viscosity.

[ScottHorn]

Is the reaction vessel you're pumping the glycerol into pressurized? If not, you'd be a lot better off with a peristaltic pump. I don't know the inner workings of the HP1050, but if nothing else I would expect the check valves to have a hard time seating.

[James_Ball]

The linearity problem with the flows is probably a factor of how the pump calculates flow since it is calibrated for solvents of much lower viscosity. What you may want to do is build your own calibration curve based on set flow versus actual flow and apply the correction to figure out where to set the flow on the software.

Is there any way you could place the entire pump into a box and elevate the temperature to at least 30-35C? That way you would reduce the viscosity throughout the whole system a little.

[jyclavier]

Did you finally solve your problem with heating solution ?
More generaly, I would be curious to know up to which value of viscsity the pump would work
Glycerol is about 1.3 Pa.Sec
I would be interested to pump a (hot) mixture 1/10 of this.

[Peter Apps]

Hello,

First, let me say the title may be a bit misleading. I am not doing HPLC, but I am trying to use an HP1050 HPLC pump to pump glycerol into a reaction system. The pump works fine using water, methanol, etc but is having some problems pumping glycerol. I am trying to flow at 0.333 ml/minute into a system at vacuum. At first, I tried heating the glycerol in the solvent bottle to help lower the viscosity, to no avail. The high surface area/volume ratio of the smaller diameter tubings caused the sample to cool down before it even reached the piston block.

Then, I thought the problem may have been that the tubing was unable to deliver glycerol to the first piston, even if the first piston was pulling full vacuum pressure in the piston chamber. I addressed this by pressurizing (in a rated pressure vessel) the headspace of the glycerol feed bottle to 160 psi. This allowed the pump to flow some glycerol, but eventually the pump maxed out on pressure.

To address the maxing out on outlet pressure, I removed the glass filter frit on the outlet valve, and it brought the pressure down to a reasonable level.

Then, I noticed the pumping rate was not linear with respect to indicated pumping rate. For instance, at 0.5 ml/min the pump put out a droplet of glycerol every 5.5 seconds. At 0.25 ml/min the pump put out a droplet every 7.5 seconds. The droplets have roughly the same volume, and are a good estimate of volumetric flow.

Tl;dr : Does anyone have any suggestions on how to use this pump with a very viscous solution such as glycerol, and get a good correspondance between indicated flow/actual flow? Thanks.

You are making life difficult for yourself. You have a vacuum that you want a viscous liquid to flow into - just use wide enough tubing, connectors and a needle valve to give the flow you need.

Peter

[bunnahabhain]

You are making life difficult for yourself. You have a vacuum that you want a viscous liquid to flow into - just use wide enough tubing, connectors and a needle valve to give the flow you need.

Peter

If the vacuum is not constant, you won't be able to get constant flow using this approach. The best way to go would be a good old fashione peristaltic pump.

Jörg

[vickig]

or a syringe pump

[Sallybeetle]

I admire your ingenuity.

[SURESH R.S]

Hi,

Usually to pump high viscous solvents HPLC pumps are not used. Further, there two parameters viz, compressibility and refill speed in the HPLC pump. Depending on the solvent pumped they need to be changed. I am not sure the pump/software you are using can take these two parameters.
You can use a infusion pump instead,

SURESH R.S