Archive for the ‘physical chemistry’ Category

Exploring the mutual solubility of phenol and water at the Faculty of Pharmacy, National University of Malaysia.

The phenol-water system is a well-studied example of what physical chemists call partially miscible liquids. The extent of miscibility is determined by temperature, as can be seen from the graph below. The inverted U-shaped curve can be regarded as made up of two halves, the one to the left being the solubility curve of phenol in water and the other the solubility curve of water in phenol. The curves meet at the temperature (66°C) where the saturated solutions of water in phenol, and phenol in water, have the same composition.

The thermodynamic forces driving the behavior of the phenol-water system are first visible in the upwardly convex mutual solubility curve, showing that the enthalpy of solution (ΔHs) in the saturated solution is positive i.e. that the system absorbs heat and so solubility increases with temperature in accordance with Le Châtelier’s Principle.

More rigorously, one can ascertain whether solubility of the minor component increases or decreases with temperature by computing

where s is the solubility expressed in mole fraction units, f2 is the corresponding activity coefficient and Ts is the temperature at which the solution is saturated.

The line of the umbrella curve charts the variation in composition of saturated solutions – phenol in water and water in phenol – with temperature. The area to the left of the curve represents unsaturated solutions of phenol in water and the area to the right represents unsaturated solutions of water in phenol, while the area above the curve represents solutions of phenol and water that are fully miscible i.e. miscible in all proportions.

But what about the area inside the curve, which is beyond the saturation limits of water in phenol and phenol in water? In this region, the system exhibits its most striking characteristic – it divides into two coexistent phases, the upper phase being a saturated solution of phenol in water and the lower phase a saturated solution of water in phenol. The curious feature of these phases is that for a given temperature their composition is fixed even though the total amounts of phenol and water composing them may vary.

To analyze how this comes about, consider the dotted line on the diagram below, which represents the composition of the phenol-water system at 50°C.

Starting with a system which consists of water only we gradually dissolve phenol in it, maintaining the temperature at 50°C, until we reach the point Y on the curve at which the phenol-in-water solution becomes saturated.

Now imagine adding to the saturated solution a small additional amount of phenol. It cannot dissolve in the solution and therefore creates a separate coexistent phase. Since this newly-formed phenol phase contains no water, the chemical potential of water in the solution provides the driving force for water to pass from the aqueous solution into the phenol phase. This cannot happen on its own however since water passing out of a phenol-saturated solution would cause the solution to become supersaturated. This would constitute change from a stable state to an unstable state which cannot occur spontaneously.

What can be postulated to occur is that the movement of water from the solution into the phenol phase simultaneously lowers the chemical potential of phenol in that coexistent phase, allowing phenol to move with the water in such proportion that the phenol-in-water phase remains saturated – as it must do since the temperature remains constant. In other words, saturated solution passes spontaneously from the aqueous phase into the phenol phase, diminishing the amount of the former and increasing the amount of the latter. Because water is the major component of the phenol-in-water phase, this bulk movement will continuously increase the proportion of water in the coexistent water-in-phenol phase until it reaches the saturation point whose composition is given by point Z on the mutual solubility curve.

In terms of chemical potential in the two-phase system, equilibrium at a given temperature will be reached when:

upper phase = sat. soln. of phenol in water
lower phase = sat. soln. of water in phenol

– – – –

Applying the Phase Rule

F = C – P + 2

First derived by the American mathematical physicist J. Willard Gibbs (1839-1903), the phase rule computes the number of system variables F which can be independently varied for a system of C components and P phases in a state of thermodynamic equilibrium.

Applying the rule to the 1 Phase region of the phenol-water system, F = 2 – 1 + 2 = 3 where the system variables are temperature, pressure and composition. So for a chosen temperature and pressure, e.g. atmospheric pressure, the composition of the phase can also be varied.

In the 2 Phase region of the phenol-water system, F = 2 – 2 + 2 = 2. So for a chosen temperature and pressure, e.g. atmospheric pressure, the compositions of the two phases are invariant.

In the diagram below, the compositions of the upper and lower phases remain invariant along the line joining Y and Z, the pressure being atmospheric and the temperature being maintained at 50°C. As we have seen, the upper layer will be a saturated solution of phenol in water where the point Y determines the % weight of phenol (= 11%). Correspondingly, the lower layer will be a saturated solution of water in phenol where the point Z determines the % weight of phenol (= 63%).

– – – –

Relationship between X, Y and Z

If a mixture of phenol and water is prepared containing X% by weight of phenol where X is between the points Y and Z as indicated on the above diagram, the mixture will form two phases whose phenol content at equilibrium is Y% by weight in the upper phase and Z% by weight in the lower phase.

Let the mass of the upper phase be M1 and that of the lower phase be M2. The mass of phenol in these two phases is therefore Y% of M1 + Z% of M2. Conservation of mass dictates that this must also equal X% of M1 + M2. Therefore

The relative masses of the upper and lower phases change according to the position of X along the line Y-Z. As X approaches Y the upper phase increases as the lower phase diminishes, becoming one phase of saturated phenol-in-water at point Y. Conversely as X approaches Z the lower phase increases as the upper phase diminishes, becoming one phase of saturated water-in-phenol at point Z.

– – – –

Further reading

Logan S.R. Journal of Chemical Education 1998

This paper uses the well-known thermodynamic equation ΔG = ΔH + TΔS as a theoretical basis for determining the circumstances under which spontaneous mixing occurs when two partially miscible liquids are brought together at constant temperature and pressure.

The approach involves the construction of equations for estimating both the enthalpy and entropy of mixing in terms of the mole fraction x of one component, and graphing the change in Gibbs free energy ΔG against x to determine the position of any minimum/minima. The paper goes on to examine the criteria for the existence of two phases on the basis of determining the circumstances under which a system of two phases will have a combined ΔG value that is lower than the corresponding ΔG for a single phase.

The conclusion is reached that the assessment of ΔG on mixing two liquids can provide a qualitative explanation of some of the phenomena observed in relation to the miscibility of two liquids.

The paper is available from the link below (pay to view)

– – – –

P Mander November 2018


Wilhelmy’s birthplace – Stargard, Pomerania – in less happy times

The mid-point of the 19th century – 1850 – was a milestone year for the neophyte science of thermodynamics. In that year, Rudolf Clausius in Germany gave the first clear joint statement of the first and second laws, upon which Josiah Willard Gibbs in America would develop chemical thermodynamics. 1850 was also the year that the allied discipline of chemical kinetics was born, thanks to the pioneering work of Ludwig Wilhelmy.

– – – –

Ludwig Ferdinand Wilhelmy was born on Christmas Day 1812 in Stargard, Pomerania (now Poland). After completing his schooling, he studied pharmacy and subsequently bought an apothecary shop. In 1843, at the age of 31, he sold the shop in order to pursue research interests at university, where he made the acquaintance of Rudolf Clausius and Hermann von Helmholtz. In 1846, Wilhelmy received his doctorate from Heidelberg University, and it was here in 1850 that he conducted the first quantitative experiments in chemical kinetics, using a polarimeter to study the rate of inversion of sucrose by acid-mediated hydrolysis.

Wilhelmy’s work had a seminal quality to it because – apart from being a talented individual – he observed that great guiding principle when commencing an exploration of the unknown: he kept things simple.

He chose a monomolecular decomposition reaction, used a large volume of water to keep the acid concentration unchanged during the experiment, maintained constant temperature and followed the inversion process with a polarimeter, which did not physically disturb the conditions of the system under study. By rigorously limiting system variables, Wilhelmy discovered a simple truth: the rate of change of sucrose concentration at any moment is proportional to the sucrose concentration at that moment.

Now it just so happened that in Wilhelmy’s earlier doctoral studies, he had become familiar with utilizing differential equations. So it was a straightforward task for him to model his new discovery as an initial value problem, which he wrote as

where Z is the concentration of sucrose, T is time, S is the acid concentration (presumed unchanging throughout the reaction), and M is a constant today called the reaction velocity constant. Wilhelmy integrated this equation to

where C is the constant of integration. Recognising that when T = 0 the sucrose concentration is its initial value Z0, he wrote


He then proceded to show that this equation was consistent with his experimental results, and thus became the first to put chemical kinetics on a theoretical foundation.


A page from Wilhelmy’s pioneering work “Ueber das Gesetz, nach welchem die Einwirkung der Säuren auf den Rohrzucker stattfindet”, published in Annalen der Physik und Chemie 81 (1850), 413–433, 499–526

– – – –

Inversion of sucrose


Sucrose has a dextrorotatory effect on polarized light, but on acid hydrolysis the resulting mixture of glucose and fructose is levorotatory, because the levorotatory fructose has a greater molar rotation than the dextrorotatory glucose. As the sucrose is used up and the glucose-fructose mixture is formed, the angle of rotation to the right (as the observer looks into the polarimeter tube) becomes less and less. It can be demonstrated that the angle of rotation is directly proportional to the sucrose concentration at any moment during the inversion process.


A Laurent polarimeter from around 1900.

– – – –

A neglected pioneer

Ludwig Wilhelmy’s groundbreaking research into the kinetics of sucrose inversion was published in Annalen der Physik und Chemie in 1850, but it failed to garner attention and in 1854 he left Heidelberg, retiring to private life in Berlin at the age of 42. He died ten years later in 1864, his pioneering work still unrecognized.

It was not until 1884, twenty years after Wilhelmy’s death and thirty four years after his great work, that Wilhelm Ostwald – one of the founding fathers of physical chemisty – called attention to Wilhelmy’s paper. Among those who took notice was the talented Dutch theoretian JH van ‘t Hoff, who in 1884 was engaged on kinetic studies of his own, soon to be published in the milestone monograph Études de dynamique chimique (Studies in Chemical Dynamics). In this book, van ‘t Hoff extended and generalized the mathematical analysis that had originally been given by Wilhelmy.

– – – –

A digression on half-life


In his study of sucrose inversion, Ludwig Wilhelmy showed that the instantaneous reaction rate was proportional to the sucrose concentration at that moment, a result he expressed mathematically as

In his published paper, Wilhelmy did not mention the fact that the fraction of sucrose consumed in a given time is independent of the initial amount. But he might well have noticed that by expressing Z as a fraction of Z0, the left hand side of the equation simply becomes the logarithm of a dimensionless number.

For example the half-life time (T0.5) i.e. the time at which half of the substance present at T0 has been consumed

JH van ‘t Hoff was well aware of this fact – he derived a half-life expression on page 3 of the Études. And in all likelihood he was also aware that this kinetic truth produced a conflict with the thermodynamic necessity for a chemical reaction to reach equilibrium.

Reconciling kinetics and thermodynamics

The starting concentration of sucrose in Wilhelmy’s inversion experiment is Z0. So after the half life period T0.5 has elapsed, the sucrose concentration will be Z0/2. After further successive half life periods the concentrations will be Z0/4, Z0/8, Z0/16 and so on. The fraction of sucrose consumed after n half lives is

This is a convergent series whose sum is Z0 – corresponding to the total consumption of the sucrose and the end of the reaction. The problem with this formula is that it implies that Wilhelmy’s sucrose inversion reaction – or any first order reaction – will take an infinitely long time to complete. This is not consistent with the fact that chemical reactions are observed to attain thermodynamic equilibrium in finite timescales.

In the Études, van ‘t Hoff successfully reconciled kinetic truth and thermodynamic necessity by advancing the idea that a chemical reaction can take place in both directions, and that the thermodynamic equilibrium constant Kc is in fact the quotient of the kinetic velocity constants for the forward (k1) and reverse (k-1) reactions

– – – –

Wilhelmy’s legacy

Wilhelmy’s pioneering work may not have been recognised in his lifetime, but the science of chemical kinetics which began with him developed into a major branch of physical chemistry, involving many famous names along the way.

In the 1880s, van ‘t Hoff and the Swedish physical chemist Svante Arrhenius – both winners of the Nobel Prize – made important theoretical advances regarding the temperature dependence of reaction rates, which proved a difficult problem to crack.

In 1899, DL Chapman proposed his theory of detonation. The chemical kinetics of explosive reactions was then taken forward by Jens Anton Christiansen, whose idea of chain reactions was further developed by Nikolaj Semyonov and Cyril Norman Hinshelwood, both of whom won the Nobel Prize for their development of the concept of branching chain reactions, and the factors that influence initiation and termination.

Several other Nobel Prize winners have their names associated with chemical kinetics, including Walther Nernst, Irving Langmuir, George Porter and JC Polanyi. The work of all these illustrious men has enriched this important subject.

But for now, we must return to Ludwig Wilhelmy in Heidelberg. It is 1854, and having failed to garner any interest in his seminal studies, he has packed his bags at the university, handed in his keys at the porter’s lodge, and is ready to begin the long journey home to Berlin.


Heidelberg, Germany in the 1850s

– – – –

Mouse-over links to works referred to in this post

Jacobus Henricus van ‘t Hoff Studies in Chemical Dynamics

Ludwig Wilhelmy “Ueber das Gesetz, nach welchem die Einwirkung der Säuren auf den Rohrzucker stattfindet” , published in Annalen der Physik und Chemie 81 (1850), 413–433, 499–526

P Mander April 2016

Credit: Microbiology Online Notes

Although the Carbon Cycle is a well-accepted concept illustrated by countless graphics on the internet such as the one shown above, I wonder if it deserves to be called a cycle in the general sense of uninterrupted circular motion. Because the fossil fuels (coal, oil and gas) formed over millions of years from atmospheric CO2 are actually end products. Left to themselves they would remain as coal, oil and gas and the cycle would stop turning. To say that the cycle is completed by human interference sounds somewhat contrived.

But despite this criticism, the Carbon Cycle has an obvious value: it helps us to see a bigger picture. And this broader understanding can be further enhanced by looking at another aspect of carbon which changes during its journey around the cycle – namely its oxidation state.

I have not come across a graphic that includes this, so here is one I drew to illustrate the idea.

Rather than describing carbon in terms of its sequence of physical transformations, this cycle shows how carbon’s oxidation state reflects changes in the nature of its chemical bonds. Oxidation states are conveniently if somewhat abstractly represented by dimensionless numbers, which in the case of carbon range from +4 (most oxidized state) to –4 (most reduced state). The lower the number, the more energy is present in carbon’s chemical bonds. So in the process of carboniferous fuel creation the oxidation state number decreases. Conversely, the process of energy release from carboniferous fuel results in an increase in oxidation state number. The natural abundance of carbon with its wide range of oxidation states centered about zero is what gives carbon its usefulness as both a source and a store of energy.

– – – –

Energy in, Energy out

Below is a quantified illustration of how carboniferous fuels store and release more energy as the oxidation state number decreases.

Energy release in kJ per mole of CO2 formed. Numbers are indicative

The green arrow shows the oxidation state decrease from +4 to 0 associated with photosynthesis in terrestrial plants and marine plankton. The carbon in the repeating molecular unit is reduced by hydrogenation, and combusting this fuel e.g. in the form of cellulose releases 447 kJ per mole of CO2 formed. Further reduction to oxidation state –1 is associated with the creation of coal (not illustrated) which releases about 510 kJ per mole of CO2 formed. Continued reduction to oxidation state –2 is associated with petroleum which releases around 610 kJ per mole of CO2 formed. Finally the formation of natural gas represents the lowest possible oxidation state of carbon, –4. On combustion natural gas releases the maximum energy of 810 kJ per mole of CO2 formed.

These numbers illustrate a general (inverse) relationship between the magnitude of the carbon oxidation state and the amount of energy generated by combustion (see Appendix 1 for more data). For each mole of CO2 released, natural gas (–4) produces nearly twice as much energy as cellulosic biomass (0). That is an appreciable difference, which perhaps deserves more attention in public discourse about greenhouse gas emissions than it receives.

It is also worth noting that although it takes millions of years for natural gas to be formed from atmospheric carbon dioxide in Nature’s Carbon Cycle, the same carbon transformation can be achieved by human beings on a vastly accelerated timescale using a process known as the Sabatier reaction. This has been recently demonstrated in a remarkable Power-to-Gas (P2G) project conducted in Austria.

– – – –

Underground storage and conversion

In Pilsbach, Upper Austria, energy company RAG Austria AG conducted a P2G project called Underground Sun Storage in which excess electricity production from wind and solar was converted by electrolysis of water to hydrogen gas which was then pumped down into a depleted natural gas reservoir at a depth of 1 km. Following the successful conclusion of this project, a second P2G project called Underground Sun Conversion was then initiated in which carbon dioxide sourced from biomass combustion or DAC was co-injected with hydrogen into the gas reservoir.

According to RAG, the pores in the matrix of the underground reservoir contain micro-organisms which within a relatively short time convert the hydrogen and carbon dioxide into natural gas, recreating the process by which natural gas originates but shortening the timescale by millions of years. In its project description RAG Austria claims that “this enables the creation of a sustainable carbon cycle”.

How the micro-organisms effect the reaction between H2 and CO2 is not described in the material I have seen. Perhaps microbial enzymes serve as catalysts – the Sabatier reaction is spontaneous and indeed thermodynamically favored under the temperature and pressure conditions of the reservoir (313K, 107 bar).

– – – –

Energy in, Energy out (again)

Ok, so let’s take a look at the thermodynamics of the processes by which RAG Austria turns carbon dioxide into natural gas. Applying Hess’s Law, it can be seen that the sum of the electrolytic process and the Sabatier reaction is equivalent to a reversal of methane combustion and corresponds to the energy stored underground in the C-H bonds of the methane molecule – note that the oxygen is formed above ground during electrolysis and is vented to the atmosphere*

*the sum of reactions bears a curious similarity to the process of photosynthesis: 6CO2 + 6H2O → C6H12O6 + 6O2

Energy loss is intrinsic to both parts of this methane synthesis program. The electrical efficiency of water electrolysis using current best practises is 70–80%, while the Sabatier reaction between H2 and CO2 taking place in the underground reservoir is exothermic and loses around 15% of the energy used to form hydrogen in the initial stage.

On the face of it, underground storage in a natural gas reservoir of hydrogen alone would seem to offer better process economics. But carbon capture and underground conversion can be titrated to achieve a variable quotient between stored hydrogen and converted methane. Both have their economic attractions; what methane lacks in terms of process inefficiencies can be compensated for in several ways in relation to hydrogen. Superior energy density, more efficient transportation as LNG and compatibility with existing energy supply infrastructures are some of them. And then there is the larger issue of the value that society places upon the desire for carbon neutrality in existing energy systems on the one hand, and the promise of carbon-free energy systems on the other.

– – – –

Appendix 1

Carbon oxidation state and heat of combustion per mole of CO2 produced

Negative numbers in the last column indicate exothermic reaction i.e. heat release. Units are kJmol-1

– – – –

Suggested further reading

EURAKTIV article on energy storage projects June 2019

RAG Austria AG website – Underground Sun Conversion

– – – –

P Mander June 2019

Credit: Extinction Rebellion

On Tuesday 2nd April 2019, environmental activists from the Extinction Rebellion group staged a parliamentary protest in London by glueing their bottoms to the windows of the viewing gallery of the House of Commons during a Brexit debate.

They used this cheeky tactic to call on British politicians to act on the ‘Climate and Ecological Crisis’. On its website, Extinction Rebellion wrote that “Government must act now to halt biodiversity loss and reduce greenhouse gas emissions to net zero by 2025”.

There is an irony here in that the call to immediate action was delivered at the precise moment when the UK government was in a state of complete paralysis over Brexit. Then again, the demonstrators’ demands weren’t meant to be taken entirely at face value. Even the most optimistic environmental biologists and chemical engineers would shake their heads at what was proposed to be accomplished in the space of just six years.

What the demonstration did achieve was to focus attention on the task in hand and the current state of progress. And it prompted CarnotCycle to pen this post on one promising technology for carbon capture and conversion (CCC) in which UK researchers are playing a key role.

– – – –

It is a well-known fact, not to mention a subject of serious environmental concern, that cows burp methane as a result of anaerobic fermentation of the grass they eat. And being aerobic respirers they also exhale carbon dioxide. Both these substances are potent greenhouse gases associated with human activity whose atmospheric levels, according to the best available science, must at least be stabilized in order to stand a chance of keeping global warming within manageable limits.

The cow in our picture is asking us an interesting question. What if it were possible to react carbon dioxide and methane together to form products that are not greenhouse gases? Even better, what if the reaction products could be put to useful purposes? Just think how cute that would be!

Putting our physical chemistry hats on for a moment and looking at the above equation, we notice that the oxidation states of carbon in the two molecules are at opposite ends of the scale. Methane has the most reduced form of carbon (-4) while carbon dioxide has the most oxidized form of carbon (+4). A redox reaction between the two looks possible and indeed is possible, albeit at elevated temperatures:

This process – called dry reforming of methane or DRM – was first introduced by Germany’s dynamic duo, Franz Fischer and Hans Tropsch, in 1928 but extensive investigation only started in the 1990s when increasing concerns about the greenhouse effect were raised by the international scientific community.

Notice how DRM simultaneously converts two greenhouse gases into two non-greenhouse gases which together make valuable syngas (1:1), a key industrial intermediate in the production of chemicals and clean fuels.

– – – –

Well, sort of. The big difficulty in making DRM viable relates to side reactions. The DRM reaction proceeds above 918K but then so does the thermal decomposition of both methane and carbon monoxide which results in carbon being deposited on the catalyst, clogging up the pores and thereby deactivating it.

This problem can however be mitigated in a very neat way by combining DRM with another methane-reforming process, namely steam reforming (SRM). Coupling reactions in this way not only reutilizes the deposited carbon but also adds a product stream with an H2/CO ratio of 3:1 which enables the syngas ratio to be adjusted for the synthesis of methanol, ethanoic acid or dimethyl ether (DME), which has promise as a sulfur-free diesel fuel, or towards the (2n+1):n H2/CO ratio required for Fischer-Tropsch synthesis of alkane fuels

The other problem is that all these reactions are endothermic (heat requiring). This energy has to be obtained from somewhere, and now here comes the next neat idea. Adding oxygen to the reactant stream allows partial oxidation of methane (POM) and catalytic combustion of methane (CCM) to take place, which are exothermic reactions that can supply the necessary heat

Putting three reforming agents – carbon dioxide, water and oxygen – together in the reactant stream with a methane feedstock creates a sufficiently energy-efficient overall process known as ‘tri-reforming’.

– – – –

To complete the conceptual scheme, the tri-reforming process is integrated into a cycle where the syngas output is utilized in the tri-generation of fuels, industrial chemicals and electricity, with the flue gases from these processes being fed back after nitrogen purging to the tri-reforming reactor. Carbon dioxide can also be fed into the cycle from external sources such as power plants and cement works.

Note that in principle, carbon dioxide can be fed into the cycle from established carbon capture and storage (CCS) processes. In this way underground reservoirs of anthropogenic carbon dioxide can be utilized as a feedstock for additional tri-generation.

– – – –

Almost. I mentioned at the start that the UK was involved in CCC research so I should say a few words about that. Scientists at Oxford and Cambridge are working with the King Abdulaziz City for Science and Technology in Saudi Arabia and the National Natural Science Foundation of China on the tri-reforming/tri-generation technology detailed above.

Since China is the world’s largest CO2 emitter and Saudi Arabia is the world’s largest oil producer, the Anglo-Sino-Saudi initiative seems a sensible geoscientific cluster. The news was announced on 28 January 2018, more than a year before the Extinction Rebellion demonstrations in London.

– – – –

Suggested further reading

A mini-review on CO2 reforming of methane

Published in June 2018 this is a useful and easily readable grounder covering the thermodynamic, kinetic, catalysis and commercial aspects of the subject.

Turning carbon dioxide into fuel

A paper co-written by the Oxbridge scientists involved in the Anglo-Sino-Saudi initiative. It was published in 2010, which shows that these guys have been on the case a while. Climate activists take note, and read their stuff.

Turning carbon dioxide into fuel – a new UK-China-Saudi Arabia initiative

The January 2018 press release referred to above. I reckon my poly-alliterative Anglo-Sino-Saudi sounds better.

Tri-reforming: a new process for reducing CO2 emissions

A bedrock paper (January 2001) from Chunshan Song at Penn State. The process diagram featured above is taken from this paper. If you don’t read anything else, read this one.

– – – –

P Mander May 2019


I dare say most of you will remember this classroom demonstration, in which water passes through a semi-permeable membrane and causes the liquid level to rise in the stem of the thistle funnel. The phenomenon is called osmosis, and at equilibrium the osmotic pressure is equal to the hydrostatic pressure.

Historical background

This experiment has its origins way back in the mid-18th century, when a French clergyman named Jean-Antoine Nollet tied a piece of pig’s bladder over the mouth of a jar containing alcohol and immersed the whole thing in a vat of water. What prompted him to do this is not known, but we do know the result of his experiment. The bladder swelled up and ultimately burst from the internal pressure.


Jean-Antoine Nollet 1700-1770

Nollet published his findings in Recherches sur les causes de Bouillonement des Liquides (1748) in which he gave a correct interpretation of the phenomenon, which arises from the much more marked permeability of the bladder to water as compared with alcohol.


Moritz Traube 1826-1894

The actual measurement of osmotic pressure had to wait for over a century, until the German chemist Moritz Traube showed in 1867 that artificial semipermeable membranes could be made using gelatin tannate or copper ferrocyanide. Traube’s compatriot Wilhelm Pfeffer, a botanist, succeded in depositing the latter in the walls of a porous jar, which when filled with a sugar solution, connected to a mercury manometer and then plunged into pure water, provided a means of measuring osmotic pressures.


Wilhelm Pfeffer 1845-1920

Following Pfeffer’s osmotic pressure measurements using sucrose solutions, on which JH van ‘t Hoff based his famously flawed gaseous theory of solutions, there were two notable teams of experimentalists – one on each side of the Atlantic – which provided high quality osmotic pressure data to test the ideas of theoreticians. In the USA, Harmon Northrop Morse and Joseph Christie Whitney Frazer led a team at Johns Hopkins University, Baltimore, Maryland from 1901 to 1923. In Britain meanwhile, the aristocrat-turned-scientist Lord Berkeley and co-worker Ernald Hartley set up a private research laboratory near Oxford which operated (with gaps due to war service) from 1904 to 1928.


Ernald Hartley (1875-1947) Besides being a research chemist, Hartley was an amateur clarinetist who played in the Oxford orchestra for many years. The photo dates from 1925.

While Morse and Frazer used the same principle as Pfeffer, albeit with a more advanced electrochemical method of depositing the membrane in the pores, Berkeley and Hartley reversed the arrangement of solvent and solution, applying measured pressure to the latter to attain equilibrium.


– – – –

Theoretical development in Europe

In Europe, the rigorous application of thermodynamics to the phenomenon of osmosis started in 1887 with Lord Rayleigh, who combined the use of the ideal gas law PV = nRT with the idea of a reversible isothermal cycle of operations in which the sum of work in the complete cycle is zero.


Lord Rayleigh 1842-1919

Being essentially an attempt to provide hypothesis-free support to van ‘t Hoff’s troubled gaseous theory of solutions, the solute in Rayleigh’s cycle was a mole of ideal gas, which was first dissolved in the solution by applied pressure and then recovered from the solution by osmotic pressure to return the system to its original state.

Rayleigh’s approach, using a zero-sum cycle of operations, was thermodynamically sound and continued to form the basis of theoretical development in its next phase, which in Europe focused on vapor pressure following the influential papers of Alfred Porter in 1907 and Hugh Callendar in 1908.


Alfred Porter 1863-1939

By 1928, the theoretical model in JAV Butler’s popular textbook The Fundamentals of Chemical Thermodynamics was close to the familiar classroom demonstration of osmosis shown at the head of this post, in which the hydrostatic pressure acting on the solution counteracts the tendency of the solvent to pass through the semi-permeable membrane. At equilibrium, the hydrostatic pressure P is equal to the osmotic pressure.


JAV Butler 1899-1977

To obtain a thermodynamic relation for osmotic pressure in terms of vapor pressures, Butler uses Rayleigh’s idea of a reversible isothermal cycle of operations together with a semipermeable membrane in the form of a movable piston between the solution and the solvent:


The diagram shows a solution under hydrostatic pressure P which is equal to the osmotic pressure. Below the semi-permeable piston is pure solvent. Butler then applies the following argumentation:

1] Vaporize 1 mole of the pure solvent at its vapor pressure p0, and expand it reversibly so that the vapor pressure falls to p equal to the partial pressure of the solvent in the solution (Butler assumes that p is not affected by P applied to the solution). Condense the vapor into the solution. Since the work of vaporization and condensation cancel out, the only work done is the work of expansion. Assuming the vapor obeys the ideal gas law, the work (w) done is given by the textbook isothermal expansion formula

2] Now move the semi-permeable piston up against the pressure P until a quantity of solvent equivalent to 1 mole of vapor has passed through it. If the decrease in the volume of the solution is ΔV, the work done is PΔV.

The cycle is now complete and the system has returned to its original state. The total work done is zero and we may equate the two terms


where P is the osmotic pressure, ΔV is the partial molal volume of the solvent in the solution, p0 is the vapor pressure of the pure solvent and p is the vapor pressure of the solvent in the solution. This thermodynamically exact relation, which involves measured vapor pressures, is in good agreement with experimental determinations of osmotic pressure at all concentrations.

There is a great irony here, in that this equation is exactly the one that JH van ‘t Hoff found his way to in Studies in Chemical Dynamics (1884), before he abandoned his good work and went completely off-track with his gaseous theory of solutions.


JH van ‘t Hoff 1852-1911

– – – –

Theoretical development in America

In the US, the theory of the semipermeable membrane and the ‘equilibrium of osmotic forces’ was the work of one supremely gifted man, Josiah Willard Gibbs, who more or less single-handedly laid the theoretical foundations of chemical thermodynamics in his milestone monograph On the Equilibrium of Heterogeneous Substances.


J Willard Gibbs 1839-1903

But before delving into the powerful idea he introduced, let us return to the subject of equilibrium in a system subject to osmotic pressure with a set-up that is slightly different to that used by Butler. In the diagram below, the piston supplies pressure Psoln to the solution which is just enough to stop solvent passing through the membrane and bring about equilibrium at constant temperature; the osmotic pressure is defined as the excess pressure Psoln – p01.


The question can now be asked: Does the condition of osmotic equilibrium coincide with equality of a thermodynamic variable on either side of the membrane? Clearly it cannot be pressure or volume, nor can it be temperature since constant temperature does not prevent osmotic disequilibrium.

The P, V, T variables do not provide an affirmative answer, but in his monumental masterwork, Gibbs supplied one of his own invention which did – the chemical potential, symbolized μ. It is an intensive variable which acts as a ‘generalized force’, driving a system from one state to another. In the present context the force drives chemical components, capable of passing through a membrane, from a state of higher potential to a state of lower potential.

So given a membrane dividing solution from solvent and permeable only to the latter, we can understand the osmotic force driving the solvent (designated by subscript 1) through the membrane into the solution in terms of movement to a region of lower potential since

Now the difference in potential can be calculated according to the textbook formula


where x1 is the mole fraction (<1) of the solvent in the solution. To achieve equilibrium, the chemical potential of the solvent in the solution must be increased by the amount –RTlnx1 (a positive quantity since lnx1 is negative). This can be done by increasing the pressure on the solution since

is always positive (V1 is the partial molar volume of the solvent in the solution).

The osmotic pressure is defined as the excess pressure Psoln – p01. As can be seen from the diagram below, this is the pressure required to raise the chemical potential of the solvent in the solution so that it becomes equal to the chemical potential of the pure solvent.


Since the slope is V1, it follows that


Combining (2) and (3) and designating the osmotic pressure by P gives the desired equilibrium relation

This is exactly equivalent to equation (1) derived by Butler, since by his terminology

The two methods of proof are thus shown to be equivalent – we can regard osmotic pressure as the excess pressure required to increase either the chemical potential or the vapor pressure of the solvent in the solution. But Gibbs saw an advantage in using potentials, which he voiced in an 1897 letter to Nature entitled Semi-Permeable Films and Osmotic Pressure:

“The advantage of using such potentials in the theory of semi-permeable diaphragms consists … in the convenient form of the condition of equilibrium, the potential for any substance to which a diaphragm is freely permeable having the same value on both sides of the diaphragm.”

– – – –