i'm working as a diy bio. I'm finding a way to create a micro glass capillary for picking up single cells. I see this video on youtube and would like to know what is the minimum I/O diameter if I pull a capillary like this. Can it be 10-50 µm at least ?

  • 2
    $\begingroup$ What kind of cells do you want to pick up? From what sample? Do you absolutely need single cells? $\endgroup$
    – nico
    Commented Jun 30, 2013 at 8:05
  • $\begingroup$ I mean what is the minimum I/O diameter if i pull a glass capillary like that video. I'm working on single cell, inject sperm to egg ... $\endgroup$ Commented Jun 30, 2013 at 11:36
  • 1
    $\begingroup$ Well, depends what type of capillary you use (there are many different ones), how close you go to the Bunsen, how fast you pull, how long you leave the capillary on the fire etc. etc. The best thing would be to invest in a pipette puller, it is very difficult to establish. Also, for microinjection you will also need a micromanipulator (unless you have an incredibly steady hand). $\endgroup$
    – nico
    Commented Jun 30, 2013 at 11:54
  • $\begingroup$ well, i'm an IT engineer so i would like to create a micromanipulator. I don't want to purchase for a puller because it will cost too much. I just want to finding a simply way to have one with minimum cost as possible. I wonder that if i pull by hand with fastest speed i will have a capillary with 10-50 µm diameter at least. $\endgroup$ Commented Jun 30, 2013 at 13:25
  • $\begingroup$ Unless you are a skilled machinist (or know one), it's going to be very difficult to make a micromanipulator from scratch. $\endgroup$
    – jonsca
    Commented Jul 1, 2013 at 2:39

3 Answers 3


It is very difficult to establish what kind of tip you will get from pulling a capillary on a Bunsen burner. There are many variables to consider, including the glass you are using, the temperature you reach before pulling, the pulling strenght, speed and time, whether you pull once or make multiple pulls etc.

If you do not want to invest into a micropipette puller (they are indeed VERY expensive) you could at least build some mechanism to obtain a constant pull in terms of time and force.

The following paper may be of help:

An Instrument for Controlled, Automated Production of Micrometer Scale Fused Silica Pipettes

It seems you would need a fairly high speed (0.8-1 m/s) to get a tip size of <50um.

I could not really find a specific reference for pulling over a Bunsen.


Here is a reference for pulling quartz glass pipettes using a custom oxy hydrogen burner. Here is a video of a commercial burner based pipette puller. Not sure what the specs are. I guess if you can melt your glass using a filament it could be easier than using a torch (a lightbulb circuit is easier to control than flame). People move to flame when dealing with high melting point quartz.

You can get a micropipette puller from Sutter for $1500 off of Ebay. I'm not completely sure how they work mechanically, but I have used many models. Someday I'd like to look inside them (the true mechanism is underneath, inside the box). They use a metal "boxcar shaped" filament that fits around the cappilary, and this heats up when a current is passed through. Two pulleys pull at the pipette, attached to something below doing the actual pulling.

I think there might be a spring-loaded mechanism for achieving the pull, rather than a motorized linear stage moving so fast. Maybe a stage or other actuator below pre-stretches a spring so that it is pulled to different degrees during melt cycle?


Its actually not that hard however its tough to make them consistently my hand. I have a capillary puller in my lab which can give me +-.5um ID we use for a patch clamp setup but When I do it by hand [just like the video] I can get within 10-50um ID. Usually it takes two pulls first to get the elongated cone shape and the second to pull it so thin that it looks like a wire and the orifice has closed. However, if you hook up the pulled capillary via flexible plastic tubing to a syringe you can try to VERY SLOWLY get a droplet and you'll know you're good. Imaging the tip is also a challenge sometimes. There's a wired lens effect with the borosilicate I'm using


You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .