Molecular packing and friction

By Mark Devlin, Contributing Editor | TLT Lubrication Fundamentals May 2026

Let’s think about how molecules fill a contact zone and what happens to them when energy is added to the system.

---

---

I promised no complicated equations, and I plan to stick to that promise. However, if we dive into lubrication theory, we can get into a rabbit hole of named concepts and equations. From Newton to Euler to Bernoulli to Navier and Stokes to Reynolds,¹ many physicists have created equations relating flow of molecules to principles of conservation of energy, conservation of momentum and inertia. Solving these equations requires making many assumptions about the system in which molecules are flowing and many notebook pages in the trash with incorrect solutions to various partial differential equations. Instead of trying to explain these equations I want you to imagine how molecules fill a contact zone and what happens to them when energy is added to the system. 

First, I’ll return to my high school physics teacher who taught us to always perform a unit analysis when we are in doubt. To define the number of fluid molecules in a contact zone, we need to know the density of the fluid (kg/m³) and the volume (m³) of the contact zone. (Also, molecular weight to convert grams of material to moles, but let’s just stick to mass.) The surfaces in a contact zone have a relative motion which is defined by an average velocity (m/s) and shear rate caused by the difference in velocity of the sliding surfaces, which means we need to know the distance between surfaces (m). Energy is a force acting over a distance (m) and force is mass (kg) times acceleration (m/s²). With all these common units (kg, m and s) related to different physical properties, we should be able to connect fluid density (kg/m³) to dynamic viscosity (kg/m*s). We should also be able to do this as pressure (force per area) and temperature (which affects the kinetic energy of molecules in a system) vary. And why are we concerned about relating density to viscosity? The strength and number of intermolecular interactions will control how much force is needed to move molecules relative to one another (viscosity), and intermolecular interactions are related to density.

From your homework you will recall that crude oils are distilled, dewaxed, hydrotreated, hydrocracked or synthesized to get the desired molecules and molecular weights to be used as lubricants.² The base oils are then classified into American Petroleum Institute (API) groups based on sulfur (S) content, saturates content (molecules with no double bonds in the carbon [C] backbone) and viscosity index (VI). Additional analytical techniques can be used to further determine the number of molecules that do and do not have ring structures. For the molecules that do not contain rings (paraffins) the number of molecules that have branches (iso-paraffins) or do not have branches (normal-paraffins or “waxes”) can be determined. For the ring structures, do they have alternating (conjugated) double bonds (aromatics) or no double bonds (naphthenics or cycloparaffins), or are the ring structures more complex with multiple aromatics groups or cyclo-paraffins fused to one another? In general, API Group I base oils have the most aromatic and naphthenic (cycloparaffinic) molecules. API Group II base oils have more ring structures than API Group III base oils and polyalphaolefins (PAOs), which are API Group IV base oils, that contain no ring structures. As the number of ring structures in the base oil decreases, the number of iso- and normal paraffins increases since the number of molecules has to add to 100% (sorry for stating the obvious).

Now that I have made you think about abstract concepts like partial differential equations and idealized molecular structures, let me make things a little easier. Everyone has experienced having to pack up all of their worldly possessions for a move. Imagine we are packing up a kitchen and have a lot of dissimilar items: plates, cups and utensils. You don’t want to carry too many boxes, so you want to pack things as tightly as possible in each box. The number of items in a box depends upon the size of a box and the shape of the items in a box, and there will always be some space left in a box. Once boxes are packed and I put them in my car, I may need to squeeze a few of them next to each other, and a force will be applied to the sides of the boxes pushing everything in the boxes closer together (pressure). As the boxes are shaken as I carry them to my car (adding some kinetic energy) some of the items may come out of the top of the box but hopefully not the bottom of the box. The changes in the position of the plates, cups and utensils depend upon on how close they are to one another (intermolecular interactions) and the strength of the interactions (I forgot to clean everything, so there is some sticky syrup left on the plates from breakfast). The change in packing due to pressure (squishing the items in the box) or kinetic energy (causing items to push out of the box) will change the density of the various items in the box. I have cheated a little bit in describing density changes. The number of items in each box does not change, so I had to describe changes in the size of the box or that items move outside the box so that the volume occupied by the dishes, cups and utensils changes.

In a laboratory a convenient way to measure the effect of temperature and pressure on density is to have a constant mass of fluid in a variable-volume sample cell and measure changes in volume as pressure and temperature are applied. (I really didn’t cheat too much in my moving example.) In these experiments we can measure the change in density with pressure and temperature. Just like the boxes of kitchen items pressure causes density to increase (sample cell volume decreases in the experiment). Temperature causes density to decrease (sample cell volume increases in the experiment). The increase in density with an increase in pressure is the compressibility of a fluid, and the decrease in density with an increase in temperature is the thermal expansivity of a fluid. Internal pressure or the relative strength of intermolecular interactions in the fluid is related to the ratio of thermal expansivity and compressibility.^3-6 The higher the internal pressure the higher the attractive forces between molecules.

We can then look at the different API base oil groups and see trends in these thermodynamic parameters. PAOs are more compressible than Group III base oils, which are more compressible than Group II base oils, and Group I base oils are the least compressible. We can convert these observations to correlations to molecular structures with compressibility decreasing when there are more ring structures in the oil. Changes in base oil molecular structure do not affect thermal expansivity too much, but the size (molecular weight) of base oil molecules does affect thermal expansivity. Finally, internal pressure varies with API base oil group in the following way: Group I > Group II > Group III > PAO. Having ring structures in the base oil increases the strength of intermolecular interactions and decreases the ability of the base oil to be compressed. In the mini-traction machine (MTM), friction of thin base oil films at various temperatures and pressures has been measured.⁶ Friction increases as internal pressure increases and the number of ring structures in base oils increases.⁶ This connection allows tribologists to change their thinking from partial differential equations relating physical properties to friction, to thinking about friction in terms of packing of molecules and the strength of intermolecular interactions.

Of course, lubricants do not contain just base oil molecules. Polymers are the most common additive used to control viscosity at different temperatures (see upcoming homework assignment, Ref. 7). Polymers are added to lubricants at very low concentrations but can dramatically change the packing of molecules under pressure since they are much larger and more flexible than ring structures in base oils.⁸ Fluid compressibility generally increases with the addition of even small amounts of polymer to a base oil. Thermal expansivity and internal pressure can increase or decrease depending upon the concentration and type of polymer added to the base oil. Other molecules present in fully formulated lubricants such as detergents and dispersants, which control deposits, can also affect packing of the molecules in a lubricant and thus compressibility, thermal expansivity and internal pressure. The ability to connect molecular packing to viscosity and friction has opened up the ability to perform molecular modeling calculations to estimate tribological properties.⁹ No need to handle any difficult math when the computer can do this for us. 

So where do we stand after several months in the current TLT Lubrication Fundamentals series? Tribology is very beneficial to society and individuals especially if we concentrate on the efficient use of energy. Viscosity is the most critical property of a lubricant that allows tribologists to balance efficiency and protection of moving parts. Viscosity can be directly related to molecular structures in lubricants if we concentrate on how these molecules pack under the different environments (temperature, pressure and shear) that exist in contact zones. The homework this time is a reminder that it is not easy to estimate viscosity under different operating conditions, but tribologists have lots of experimental and modeling techniques to determine the rheological properties needed to design new lubricants.⁷ In the next TLT Lubrication Fundamentals article, we will discuss what occurs when surfaces are touching and the physics and chemistry tribologists use to control “boundary lubrication regime” friction.

REFERENCES
1. Gresham, R.M. (2014), “So who was this Osborne Reynolds guy, anyway?,” TLT, 70 (9), pp. 22-23. Available at www.stle.org/files/TLTArchives/2014/09_September/Lubrication_Fundamentals.aspx.
2. Holdmeyer, D. (2022), “Crude oil to base stocks to base oils to lubricating oils,” TLT, 78 (8), pp. 24-28. Available at www.stle.org/files/TLTArchives/2022/08_August/Lubrication_Fundamentals.aspx.
3. Dickmann, J. S., Devlin, M. T., Hassler, J. C. and Kiran, E. (2018), “High pressure volumetric properties and viscosity of base oils used in automotive lubricants and their modeling,” Industrial & Engineering Chemistry Research, 57, pp. 17266-17275.
4. Grandelli, H. E., Dickmann, J. S., Devlin, M. T., Hassler, J. C. and Kiran, E. (2013), “Volumetric properties and internal pressure of poly(α-olefin) base oils,” Industrial & Engineering Chemistry Research, 52 (50), pp. 17725-17734.
5. Devlin, M. T., Grandelli, H. E., Dickman, J. S., Hassler, J. C. and Kiran, E. (2016), “Volumetric and rheological properties of base oils at high pressures,” 71st STLE Annual Meeting, May 2016, Las Vegas, Nev.
6. Spikes, H. A. (1990), “A thermodynamic approach to viscosity,” Tribology Transactions, 33 (1), pp. 140-148.
7. McGuire., N. (2021), “Calculating lubricant viscosity versus temperature,” TLT, 77 (3), pp. 32-36. Available at www.stle.org/files/TLTArchives/2021/03_March/Lubrication_Fundamentals.aspx.
8. Dickmann, J. S., Avery, K., Devlin, M. T. and E. Kiran (2020), “High-pressure density, viscosity, and modeling of mixtures of a poly(α-olefin) base oil lubricant with polymeric additives,” Industrial & Engineering Chemistry Research, 59 (16), pp. 7926-7942.
9. McGuire, N. (2021), “Progress in tribological molecular simulations,” TLT, 77 (6), pp. 48-61. Available at www.stle.org/files/TLTArchives/2021/06_June/Cover_Story.aspx.