In a charming letter published in 1774, Benjamin Franklin described an experiment in which he had poured “not more than a teaspoonful” of oil onto a pond, which he found sufficient to calm the water despite a choppy wind. What, you may be wondering, connects Franklin, ponds, tears, and salad dressing? In the last “What is Biophysics?” post, I asked how your eyes’ tears are like ketchup. This is a companion piece, highlighting that even a very common and seemingly simple biological material — tears — makes use of many subtle physical forces. I pointed out last time the shear-thinning character of tears — how they flow more easily the more they’re pushed. I’ve known about that for years. Quite recently, however, I learned of another important aspect of the eyes’ liquid coating; my enlightenment was triggered by problems with my own eyes.
My eyes sometimes water a lot, especially when I’m bicycling. (Since I bike to work, this is a frequent hassle.) This started a few years ago, and strong episodes come and go. An ophthalmologist diagnosed the problem: clogged ducts. Not the tear ducts, but rather the openings of the “Meibomian Glands” that line the eyelids and secrete oil, whose existence I had been totally unaware of. Oil and water separate, just as they do in a vinaigrette salad dressing. As Franklin observed, a little bit of oil will situate itself the interface between water and air, making a thin film. This arrangement is, in brief, a consequence of water molecules’ very large affinity for each other and the relatedly very large energy associated with an air-water interface, which is ameliorated by its replacement by an air-oil and oil-water interface.
Like Ben Franklin’s pond, the watery film of the eye is coated by a thin layer of oil cast upon it. The oil layer slows evaporation. Without enough oil, the tear film easily dissipates. Sensing this, the body produces more tears — hence my watery eyes. The clogs can be exacerbated by bacteria, and bout of this a few months ago was helped by antibiotic eyedrops. (This intersects other interests of mine, but not in any way that I can offer useful insights into.)
About a century after Benjamin Franklin poured oil on a pond, Lord Rayleigh calculated the thickness of the oil layer — the oil volume and pond area are noted in Franklin’s letter, and it seems not to have occurred to Franklin to divide the one into the other. (Rayleigh made his own measurements as well.) The thickness is about a nanometer, which we now know to be the size of a single oil molecule! The layer is a molecular monolayer, in modern parlance. (All this is described in many places, for example, “Benjamin Franklin, Philadelphia’s Favorite Son, was a Membrane Biophysicist” by D-N Wang et al., Biophysical Journal 104:287-291 (2013); Link.)
I recently found across a paper (or more accurately, the abstract of a paper, since I don’t have access to the full text), “Direct Visualization of Continuous Meibum Secretion From the Orifices of Meibomian Glands to the Tear Film,” in which the authors measure the rate of oil secretion (B-J Cho et al., Cornea 38: 1245-1252 (2019); Link). In normal subjects, it’s about 3 picoliters per second, or about 3 x 10-9 cm3 / second. Given that the exposed part of the eye is about 6 cm2, and there’s about 3 seconds on average between blinks, this gives us about a 1 nm oil layer — a single molecule thick — just like Ben Franklin’s pond about a 0.015 nm thick layer of oil amassed between blinks — considerably smaller than a single molecule. I’m not sure what this means for the eyes — perhaps the complete oil layer needn’t be replenished between blinks? (Thanks to reader MM for catching my embarrassing arithmetic mistake!)
Advertisement: As noted last time, tears only pop up in my pop-science biophysics book in the context of a teaching activity gone bad, but here are the usual links anyway: My description, Publisher, Amazon.)
Today’s illustration: A bottle. (Not oil and water.) I like how the glass turned out; the liquid less so.
— Raghuveer Parthasarathy; July 14, 2022
Very interesting!
But, how do you get 1nm from those?
3 [s] * (3n) * (c[m])^3 / (6 (c[m])^2) = 1nc[m] = 0.015nm
Forgot a 1/[s] factor.
Thanks! I thought I checked this, but you’re correct — I’m off by a factor of 100! I have updated the post. I’m not sure what this means for the eyes… I suppose the complete oil layer needn’t be replenished between blinks. In any case, thank you for catching this!
Building on MM’s post and my update: I’ve now gotten the full paper from Cho et al, rather than just the abstract, thanks to Interlibrary Loan. Despite stating a volume secretion rate, the authors don’t actually measure volume — they measure the area of the secreted oil and then *assume* a thickness of 67 nm to calculate a volume! This is puzzling, and I’ll think more about it later…
That is a very particular thickness value. Wonder how they got there, as it is also a prime number.
The “67 nm” is not actually measured in this paper, and it’s given without uncertainties. The authors cite another paper; looking at that one: “In normal subjects, the median (range) LLT was 67 (33–100) nm,” so the range is quite large! Returning to the Cho et al. paper, there are aspects of the “3 pL/s” that are puzzling; at this rate and 67 nm thickness, it would take many hours for the film to cover the eyes. I’ve written to the authors for clarification. The saga continues…