Tuesday, 19 June 2018

Can curved molecules in flames help reduce soot pollution?

Soot is a serious problem which impacts human and the planet's health. The world health organisation estimates that in 2012 seven million people died or 1/8 total deaths around the world, from air pollution with around half from ambient air pollution and the rest from indoor air pollution from cooking on inefficient stoves (WHO). Soot emissions are also the biggest contributor to global warming after the greenhouse gases. So how do we stop soot from forming? The first step is to understand the mechanism by which soot forms, which is surprisingly still not yet fully understood. We have just published an article considering the impact of curved aromatic molecules on the formation mechanism of soot (here is a link to the preprint).

Image result for soot pollution
So what is currently understood about soot formation? The animation below will aid in illustrating soot formation. It shows the inside of a small lawn mower motor with a transparent plastic engine head allowing for a high-speed camera to capture the combustion inside the engine. 

A small spark ignites the petrol and air mixture and a blue flame is seen to spread through the cylinder. At the flame front there is enough oxygen and so the fuel burns completely to carbon dioxide and water with the blue colour (the colour is due to excited intermediates C2*, CH* and OH*). However, not all of the flame has enough oxygen to completely combust. Behind the flame front, a build-up of carbon molecules forms soot which glows yellow because it is heated by the combustion. So the yellow colour you see in the figure is soot which is being exhausted into the atmosphere. You can confirm that the yellow colour is from soot at home by placing a metal spoon into the yellow part of the flame and picking up black soot on the spoon.

From the animation inside the engine, you will also have seen the complexity of the flame. To simply the system we often use simple burners to study soot formation which produce very stable candle-like flames (diffusion flame). Below is one of the burners with a schematic to highlight the molecules involved in soot formation. The steps for soot formation are as follows:
  1. Precursor pyrolysis - the fuel breaks down and without enough oxygen a build up of carbon molecules - primarily acetylene occurs.
  2. PAH formation - The acetylene C2H2 reacts through a radical chain reaction into aromatic fused ring molecules.
  3. Inception/nucleation - Small carbon nanoparticle nuclei are formed 
  4. Growth - Some of these nuclei grow to become primary particles tens of nanometres in size.
  5. Aggregation/agglomeration - Primary particles stick together and form fractal-like stringy aggregates.

Step three (inception) is currently the most difficult to understand. Formation of small carbon nanoparticles from aromatic molecules in the flame happens too quickly to be a chemical reaction and must include some physical sticking or clustering to explain the speed of formation. My research group has previously determined that flat PAH (planar aromatics with only hexagonal rings) have been found to not be sticky enough to overcome the high temperatures where soot forms in a flame (~1000-1250°C). This thermal energy shakes the clusters apart.

So what about these curved PAH molecules? It is well known that they are present in soot, corannulene (a simple curved PAH with a pentagon ring surrounded by hexagonal rings) has been extracted from soot and if you decrease the pressure around the flame completely enclosed spherical cages of carbon (fullerenes) can be produced.

corannulene - Wiktionary           Fullerene C60
curved PAH corannulene (left and below click and drag to move the grey model around and zoom with your mouse wheel) and closed cage fullerene (right)

We have previously shown that these molecules contain a large electric polarisation which might be important. So how do these polarised cPAH effect soot formation? In this paper we considered three questions:
  1. What causes lead to curvature and how small do the molecules need to become curved?
  2. Can curved PAH cluster together in the flame where soot forms (~1000-1250°C)?
  3. What other interactions could cPAH have with other precursors (e.g. chemi-ions)?

What causes the curvature and how small do the molecules need to be to curve? Using computational chemistry we could calculate the geometry of lots of different curved aromatic molecules. We found the minimum size is 6 rings to curve a PAH with at least one ring being pentagonal.

It turns out the pentagon containing PAH are planar for larger molecules than you would expect from considering a perfect net of pentagons and hexagons. We found this is due to the bonding of electrons on the top and bottom of the molecule favouring planar geometries which are only overcome when the in-plane bonds are strong enough to pull the molecule into a curved configuration.

Can curved PAH cluster together in the flame where soot forms (~1000-1250°C)? Using another calculation method we worked out how sticky the cPAH are compared with their planar cousins (more negative binding energies means more strongly bound) shown below. Triangles indicate flat PAH and squares indicate cPAH. We found that for one or two pentagons, cPAH have similar binding energies to flat PAH. With three or greater number of pentagons the cPAH bound with less energy than the same sized flat PAH.
We have previously shown that for the size of flat PAH in soot it is unlikely to be able to stablise the small clusters in the soot inception zone. So for cPAH with up to two pentagons, we expect the situation to be similar and with three or more the nucleation is definitely not possible. 

What other interactions could cPAH have with other precursors? In the schematic near the top of the article, we also highlighted some ionic species called chemi-ions (coloured blue). There is a surprising amount of these charges in flames which form from reactions between an excited C-H molecule (which is one of the intermediates that glows blue). When excited CH  reacts with oxygen or acetylene it ejects an electron and becomes positively charged. The impact of these charges can be quite dramatic if you apply an electric field. At a given voltage you can also see the flame conduct electricity with arcs going through the flame. 

Soot formation can also be halted with a strong electric field. The picture below shows with a counterflow burner with fuel from below and air from above. With no electric field applied a yellow sooting flame is found. Applying a strong electric field removes the soot from the flame leaving a blue clean burning flame. This has been explained as charged nuclei and carbon being rapidly sucked out of the flame by the strong electric field.

Top - counterflow diffusion flame without an electric field applied. Bottom - with a strong electric field applied
Credit: Lawton and Weinberg 1969
Curved aromatics are significantly polar and might explain these electrical aspects of soot formation.  The interaction between charge and a dipolar molecule is long range and substantial. The plot below shows a range of curved PAH (covering the range of sizes we see in the flame 10-20 rings) with their binding energy to a charge. To give you a sense of what sort of energies are needed to hold something together at flame temperatures around 165 kJ/mol is usually quoted.

There is quite a lot more work needed to consider this ionic interaction and what impact soot formation here are some of the questions that need answering:
  • How many cPAH are present in early soot nanoparticles?
  • What is the polarity of the cPAH in early soot?
  • Can a cluster of cPAH be stabilised at high temperatures in a flame with a chemi-ion?
  • Is flexoelectricity involved in any other way in soot formation?
So the question Can curved molecules help reduce soot pollution? is yet to be fully answered. We will be presenting these results at the next combustion symposium in Dublin at the end of next month and have more results on the way.