Making the Micropipette
press.asimov.com
press.asimov.com
Engineering is all about laziness. The whole field is about making devices & structures to reduce overall labor required. Engineers need a certain sort of laziness, a willingness to work now in order to work less later (or to help others work less later). Tireless ambition rarely leads to good engineering, it leads to continued drudgery.
In addition to laziness/efficiency, curiosity and lateral thinking are helpful qualities.
There's some entropy law of abstraction that is the opposite of Occam's razor: take a simple idea and embellish it until it is sophistimicated enough to require total reengineering to decomplexify it.
If the answer is yes, consider spending a couple of days making improvements to the process, as a frustrated Schnitger did in 1957. Simple solutions to widespread problems can alter the course of scientific history.
--
sigh I guess it's the season for mucking around with my shell again
>> What is something you do, day after day, that is arduous or bothersome? Do you have a subtle itch or a pressing need to make it easier?
This is what I faced for well over 45 years of deep exposure to lab bench work from many different positions from student to grunt to leadership.
Chemical commodities, not biochemical research though.
However as expected there is a certain amount of similarity in some of the laboratory tasks regardless of the diversity of scientific objectives.
I'm still willing to grab dangerous chemicals & apparatus by the throat with my own bare hands if necessary for valuable clients. Just not wanting to use my own throat, so I never wanted to pipet by mouth either.
The vintage photo of Dr. Verder looks like she is filling a measuring pipet, before capping the glass tube with a fingertip then precisely dispensing manually as many scientists at the time would do.[0]
Now it's fire-retardant lab coats and respirator-ready facial hair as table stakes.
In my lab I had a vintage picture of Edison holding up a chemical solution above eye level, observing it carefully without safety glasses or PPE of any kind. His lab in the background was "a little bit" cluttered with chemicals and stuff. I didn't draw attenton to this kind of artwork, it was hanging off to the side in the front office, I figured it spoke for itself.
Plus I had more types of PPE than most are familiar with, which I used carefully as needed. But that's not what I'm focusing on right now.
Mouth pipetting was still common when I was getting started 30 years later than 1938, as a student so this is what had been taught for many decades but you were supposed to be a realistic judge about what risk there could be depending on what you were pipetting. The rubber bulbs which can substitute for mouth pipetting can be clumsy and not as precise and each person was expected to draw the line at what laboratory solutions they were going to avoid inhaling. These were often referred to as Pasteur pipets and were still all-glass everywhere, even thin disposable ones mostly until the 1980's when plastic disposables took over.
But by the 70's there was growing interest in safety and "rules" against mouth-pipetting were getting more widespread, but it just made people look the other way when desperate workers did it a little bit anyway, there was not usually an effective formal replacement.
Ripe for disruption by mechanical means I would guess.
Maybe with increasing complexity to the mechanism I would say it's mainly worthwhile for quite repetitive work.
Regular glass pipets are still everywhere and I was just lucky to become a professional not long before a simple precision pipet bulb was invented having little plugged vents in the rubber that are controlled by careful squeezing.[1] People from the chemical plants and refineries had never seen this, it was only available from one company until 20 years later when the patent expired. It was good for major company chemists and engineers to be able to see me manually doing NBS-traceable (now NIST-traceable) work with obvious clarity when stakes were highest. They knew their own laboratories, which supported the process units that actually made the chemicals, were way more expensively built but still somewhat embarrassing sometimes when it came to absolute certainty. Now decades later I'm not alone, "everyone" claims to be NIST-traceable, I still say not so fast. When I came to my employer almost a decade ago I had to kick out the mechanical pipets and replace them with serialized traceable manual traditional glass type, with major retraining for this exact reason. [2] Expert witnessing is more critical than routine work.
I always liked the Gilson salesman, and Ranin too but I never bought anything since the stakes are so high with the big ships, and your ship doesn't come in every day anyway so there's not that kind of repitition. But a couple barges aren't too bad either. So I've got a lifetime of ideas that other people could use where I didn't go far enough to actually even flush out a prototype since I didn't need it for myself at the time. Over the decades I've watched some of the things I had come up with, be recognized and commercialized by capable operators much later. Some of the most obvious things are the most commercializable and it was going to happen anyway. Every milestone or breakthrough doesn't have to be that elusive to be good. It doesn't bother me to be late to market when I'm not in the market to begin with. It's a good feeling knowing there's more where that came from all the time, and early ideas that nobody else has touched have now stood the test of time without competition developing. Plus if I had patented everything most of it would be expired by now.
A lot of the chemical plants only make a handful of different chemicals at each location, and usually only a few of those are actually commodities. That limited number of chemicals are the ones that each of their labs is designed and built to handle 24/7, and some tasks can get pretty repetitive for them. But when you regularly work with material from all over the place you really get ideas for inventions for all kinds of different things. Especially with emerging instruments and techniques when there aren't so many worthwhile publications to go on. So it can be a slippery slope and you end up inventing something every day. Never comercialized very much yet, too busy, a lot is not worth money anyway or would require becoming a capitalist of some kind where the terms are not very attractive. I did OK getting paid for what I could do rather than how much I sold. Well I enjoyed sales starting as a teenager and eventually starting my own commercial lab, scientific instrument sales was supposed to be a future effort but I did not approach that milestone close enough with any prototype before natural disaster set it back where you can not make up for lost time. I also had a framed vintage picture of Hewlett & Packard posing with their original invention, hanging back there.
The edge of the bench can be like a goal line where you can make field goals from a reasonable distance (if your desk or office is not too far from the bench) but you can only actually score a touchdown by crossing the line yourself. Extra points count for twice as much too when somebody personally carries the ball across the line rather than kicking it in "remotely". Plus in old labs there has never been enough space on the bench for PC's because they weren't built for it, they were just crammed in. I know. I did it myself.
I still like to build labs from the ground up, it's a hoot.
[0] In the backgroud of the photo there's a vat on a low wooden bench, that is a precision heating bath where the electronic temperature control consists of the logic necessary to turn off the heating element when the mercury in a special wired thermometer rises to make contact with a thin wire reaching into the inside of the glass from the top. The depth of the wire was adjustable with a knob at the top of the thermometer, not unlike the knob that would later appear at the top of the adjustable mechanical pipets to easily adjust their volumes. I was one of the last to use a bath of this vintage which I had gotten surplus and never thought I would ever restore. This does remind me that years later, but still before digital, when we needed a 0.001 degree tolerance for another location I rapidly designed an analog temperature controller using a solid-state sensor at the bottom of a glass tube, with the circuitry at the top in a small enclosure to run on an external bipolar power supply. I needed it to be a drop-in replacement for the glass mercury ones that were still common in the much more modern baths that were still in use. Plus it had to be built with as many parts as possible from Radio Shack, ended up the only thing I needed from elsewhere was a gigohm resistor. Another one of those everyday inventions. Once digital appeared it took years before this precision could be reached.
[1] https://www.coleparmer.com/i/silicone-pipette-filler-3-way-v... Now these are available in many polymers other than natural rubber, with openings plugged with different materials than the original stainless steel ball bearings.
[2] Also kicked out IT since they weren't very good with chemicals (or even electronics) but that's a different story altogether.
Ouch. This felt… a bit unnecessary and out of place.
> funny because it happened to a scientist that invented a tool.
..for transferring fluids.
I don’t think much of Freudian psychology as a whole, but he’s got a relevant quote that is pretty good on Gallows humor, from Der Humor 1927:
> "The ego refuses to be distressed by the provocations of reality, to let itself be compelled to suffer. It insists that it cannot be affected by the traumas of the external world; it shows, in fact, that such traumas are no more than occasions for it to gain pleasure."
The comedy, in this light, is a result of the dissonance that someone smart and savvy enough to invent an extremely useful tool for moving liquids died in a large body of liquids not long after said invention.
My guess is that people who find the humor here relate more to the drowning scientist directly. Dying from drowning in an Alpine lake, or just drowning, isn’t that common of a risk, but if it is a risk you are familiar with you’re more likely to find the humor.
At least thats the case for me.
Thank you for the explanation.
A bit of dark humor can be a nice touch.
The problem is rather if you are ok with the fact a part of your audience will be offended with it.
https://www.astralcodexten.com/p/your-incentives-are-not-the...
It is a "digital dispenser" that leverages inkjet technology to dispense complex, multi-fluid protocols with picoliter to microliter accuracy over about any type of well plate quite quickly and effectively.
The basis of the product is a one-time-use cassette that incorporates dispense heads and fluid reservoirs which the user fills with their own fluids. Depending on the model, the cassette can work with 1, 4, or 8 fluids at a time.
The latest new cassette type can sense and dispense single cells into wells with a high level of occupancy and viability.
I know this comes off as a bit of a sales pitch, but it's a product I'm genuinely excited to work on and our customers seem quite pleased with it as well. If you search for "hp digital dispenser" you'll find more info on it.
What's your precision (dispense coefficient of variation) like? When I was working with automated pipettes, we found that in order get down to single-microliter precision we had to "touch off" on the dispensing surface to use surface tension to pull the last drop off consistently. However, our aspirate/dispense ranges were pretty wide -- around 15uL to about 50mL from the same device.
It’s also no problem to do 1536 plates and even some nano-well setups using custom well plate definitions.
There went so many life science and industrial products, leaving HP as more of a PC and printer company which had registered outsized growth supplying offices and consumers during the boom. Setting the stage for smooth-talking financial manipulators to dominate the engineering culture like never before. Until it descended into more of an ink company than electronics for a while there.
So many of the brains had been at Agilent for a while by then.
IIRC the inkjet first became available in the HP3396 laboratory integrator (when most offices were still on DOS) as a built-in printer for full letter size sprocket-feed paper to print your scientific results, usually with real-time detailed graphics even if it was only black ink.
Before that, previous models used about a 4 inch wide thermal roll, the OS was outstanding and it was miniaturized like no other since HP used some of their custom application-specific chips for functions that competitive units at the time needed an entire circuit board for. Not too many times as big as a top HP calculator, the old ones had the look and feel of quality along with the performance.
Remember this is when HP was still capable of coming up with a new operating system every two years (that was more advanced than anything Microsoft had ever done).
But since the competitive integrators had a much bigger bench footprint they contained thermal printers for full letter-size output. Or 8.5 inches x more than just 11 inches on those thermal rolls or z-fold packs, since it was common to print things like long chromatograms, banner style much longer than 11 or 14 inches, resembling the output of an analog chart recorder on its rolls of graph paper. The expensive bench instruments which were well established were still all analog output which had originally fed into a dumb chart recorder. So many had a transition where the instruments were first starting to get hooked up to the analog-to-digital input of a device like this but the printouts needed to be handled in the filing cabinets (or faxed, 39 inches maximum page length) no differently than if they had come from the old chart recorders. Which were naturally maintained as backups or parallel recorders.
The little HP's could often be used as more powerful computational devices, but the printouts were more like thermal retail reciepts than the others.
So then the bigger HP3396 came out with its new inkjet doing the precision job on sprocket feed continuous letter-width "plain" paper, it was upsized with even more features. It looked so modern. Sprocket-feed was already advanced for dot-matrix with DOS computers for office work, but the inkjet on this thing really got your attention. If you were lucky you would have the optional digital storage (hard) drive which was bigger than the upsized integrator and would sit undeneath. Or maybe the less-expensive routine storage version with openings in the front for removable disks which were functionally equivalent to 5.25 inch DOS floppies, except the magnetic disk part was inside a miniaturized 3.5 inch, thin hard plastic cartridge shell thing which was not exactly flexible.
Next thing you know IBM PS/2 desktops started to appear in offices using these same type (single-sided) 3.5 inch floppies and eventually that's what PC's started to have before going double-sided.
But nobody dreamed it would be a good idea to format the disks for two different companies or usages like this in a very similar way whatsoever.
The integrators basically filled a niche before the PC's took over in the lab.
Eventually inkjets for the office and home took off, especially in color.
And the rest is history.
We had a guest once who used to be a lab nurse and mentioned using mouth pipettes. The looks of horror on the scientists in the room was palpable!
I was surprised that anyone still did it given the availability of the micropipett.
Also, I wonder why the scientists of the past didn't obtain the liquid samples by simply sticking their thumb over one end of a straw dipped in the liquid.
Brings back memories of learning how to do that with my siblings. Every chance we got, we would scrunch up straw wrappers and then use the method to animate the wrapper with liquid.
Not certain if I should show my kids how to do that.
It was terrible and nothing ever worked and my thumbs hurt.
> Radium dials were typically painted by young women, who used to 'point' their brushes by licking and shaping the bristles prior to painting the fine lines and numbers on the dials. This practice resulted in the ingestion of radium, which caused serious jaw-bone degeneration and malignancy and other dental diseases. The disease, radium-induced osteonecrosis, was recognized as an occupational disease in 1925 after a group of radium painters, known as the Radium Girls, from the United States Radium Corporation sued. By 1930, all dial painters stopped pointing their brushes by mouth. Stopping this practice drastically reduced the amount of radium ingested and therefore, the incidence of malignancy.
But being lazy one of my side projects is to make an automatical tardigrade pipettor (ML-based with motors and stuff).
The other folks pipetting by mouth are actually looking down at the tube and see the fluid volume reaches a particular well-defined line on the glass pipette. If you can taste the fluid, you've gone too far.
Pipettes are originally the glass tubes of fixed or graduated volume, but the definition has expanded to include air displacement pipettes aka "micropipettes" (even when the volume is measured in milliliters not microliters, they are still colloquially micropipettes).
The pipet is considered the tube, the pipettor is the mechanical attachment, I would say that is whether the attachment is with lips & tongue or not.
This seems ridiculously high to me. If the student works 6 days a week, 10 hours a day, they would be using a new tip about every 3-4 minutes nonstop, every single hour of those 10 hours a day 6 days a week for 5 years. No reading, no writing, no meetings - just pipette tips every 4 minutes, 10 hours a day for 5 years.
Is that really the life?
edit: Oh. Machines. Right.
Today we have machines that can batch processes using pipettes, and they can burn them at a fast rate.
When you're using 64 tips at a time, it's easy to imagine using ten boxes of 96 tips per week. Ten boxes of tips will fill that pipette only 15 times. So you're using your multipipette twice a day.
Yes it is! For example, a simple 96-well plate qPCR experiment can use 192 tips. It's easier with multichannel pipettes though.
So it's more like: use 200 tips in the morning, then do reading and meetings the rest of the day.
It's not like one pipet tip a minute. It's more like, you have N source vials and M destination vials/wells/rows, and you'll do that several times a day, interspersed with other processes. A cycle goes like:
10 New tip
20 Suck up sample from source n
30 Dispense into m-sub-i
40 while you still have destinations for the current solution, goto 20
50 eject tip
60 goto 10
This cycle takes maybe a few seconds to half a minute. Also multiply number of tips by P if you have a multi-tip pipette.
Pipetting and changing tips are a matter of seconds, not minutes.