Ozone and Photochemical Smog

See how confident you are with the Wolverines.

Okay. So we’ve talked about urban air pollution

a couple of times through this semester

and this is actually a photograph of a typical day in Beijing.

We saw the one a couple of lectures ago

when someone was trying to raise a flag in Tiananmen Square and I,

I couldn’t see the flag.

But you know following a rainfall,

Beijing can clear up and so what we’re going to talk about today

are some of the causes of urban air pollution,

specifically around smog and photochemical smog.

So what we’ll talk about today again will be

some of the primary pollutants of concern.

Criteria air pollutants as we call them

in the regulatory framework of the U.S. CPA.

We’re going to be able to do calculations using

mass balance equations just like we did earlier,

whether it was for a, you know earlier we looked at lakes and creeks merging.

We looked at reactors.

We’re going to look at an urban box model for air pollution

and we’ll do similar sorts of calculations

looking at pollutants in and out,

and those that can be generated within that volume.

We’ll talk a bit about the mechanisms of photochemical smog formation.

You may remember a couple of lectures ago

we talked about good ozone and bad ozone,

and the focus today will be on the bad ozone,

the ozone in the troposphere.

We’ll finish up with some discussion of isopleth diagrams

and ozone forming potential, and that will give you

an idea of perhaps some management tools

for looking at reducing NOx or VOCs and formation of photochemical smog.

And we’ll talk about a nice little calculation tool at the end of the lecture,

the MIR, the maximum incremental reactivity scale.

It gives us a way to look at some individual VOCs and their potential for ozone formation.

Okay so we have talked about the primary air contaminants

of concern in urban environments and criteria air pollutants.

So you may remember this figure.

So we talked about sources of air pollution.

Of course, coal burning power plants, ones that are very prevalent in China,

are responsible for emitting SOx and NOx and particulates and hydrocarbons,

and those are all criteria air pollutants.

Urban transportation, motor vehicles are also significant sources of admissions

of both NOx and hydrocarbons, particularly carbon monoxide

and other partially burned fuels.

Incinerators, other industrial stacks, can be sources of air pollution as well.

So there are a number of sources to consider.

When we do some of our box model monitor, modeling,

of course we’re going to be looking at things at a fairly large scale.

But as you reduce your scale,

if you become the environmental engineer at an industrial facility,

and someone tells you we’re going to put another stack out the roof

because we’re going to add a new process,

well then you have to look at the impact of that stack and

its emissions on the ambient air.

Are you going to be increasing NOx?

Are you going to be increasing VOCs at the property boundary?

So we talked earlier about the major criteria air pollutants,

carbon monoxide and O2 ozone, SO2, lead, and particulate matter.

The focus today will be a lot on NO2,

which is responsible for photochemical smog, the resultant pollutant ozone,

and other VOCs that are CAPs.

Another term you may hear in the air pollution vernacular

is HAPs or hazardous air pollutants.

These are typically going to be VOCs like benzene or aldehydes,

other, other volatile organic compounds, heavy metals,

and other pollutants that are emitted at much smaller doses.

But typically an air permitting authority would look at

the total sum of all the HAPs in that emission rather than individual HAPs.

So we have, we’ll start with a, a mass balance problem,

much like we did earlier in, in looking at water.

So we have a city with a given area that has a daily lead emission rate of 5,000 kg/d.

Typically, those would be from auto exhaust and other industrial processes.

Background concentrations of lead we’re going to be given at .1 mcg/m3.

Now during a hot summer day we have a mixing height of 500 meters

and a wind speed of 2 m/s.

So you’ll see when we set up the box model those are variables

that we need to look at a mass balance and a flow of air

through a given control volume.

So we have emissions of lead at 5,000 kg/d,

but we also have lead deposition from the atmosphere at .05 cm/s.

So our question is what is the steady state lead concentration in the urban air?

So remember when we started to talk about mass balance equations,

you have to get into your mind the thought that

is this steady state or non-steady state?

Is this a conservative pollutant or a non-conservative pollutant?

So with the information we have here,

everybody raise your hand who thinks this is a steady state problem?

Please more hands or else you’re going to have a rough time in two weeks.

Okay. This is a definite, we’re asking what’s the steady state lead concentration.

Second question, is this is a conservative or a non-conservative pollutant?

Conservative, raise your hands.

Okay. This, this is a conservative pollutant.

We’re not transforming lead.

We’re not degrading lead.

Now it, it’s going to be emitted and it’s going to be deposited,

and that’s going to be part of our mass balance equation,

but we’re not going to be changing the form of lead in the, in the problem.

Okay, so this is our box model.

And we we’re given the area of Leadville,

so those will be inputs into our calculations.

We set up an atmospheric mixing height that was given in the problem as 500 meters.

So we have our control volume for the air pollution problem.

It’s going to be the area of the city times the mixing height of the problem.

So we have our control volume.

We will have a wind velocity going in and out of that control volume.

We were given a pollutant emission rate M,

that was for lead in kilograms per day.

We also have a pollutant deposition velocity Vd.

So those are the, the parameters around the control volume.

So as we did before, we can set up a mass balance equation

looking at the change of mass over time being equal to the mass in minus the mass out,

plus whatever mass is reacted.

So we said this was a steady state problem,

so the dm/dt is going to be equal to 0.

Lead is not going to be transformed in this problem.

We’re not, it’s not going to be reduced or oxidized.

It’s, it’s not changing any, any forms,

so we’ll have masses going in and out as part of the mass balance,

but no reaction of lead.

So the first thing we can do is a mass balance on the air.

If you remember when we did these water problems,

the first thing we did was look at the mass balance on the water,

and the mass balance on the air is going to be pretty straightforward.

It’s going to be the flow times the density of air.

The mass of air going out is going to be the flow out times the density of air going out,

and we can set these equal to each other.

We know mass in is going to equal mass out and we can substitute a term,

rather than the flow times the density we can use the,

the velocity times the, times the cross-sectional area.

Okay, so our next step is to look at the mass balance on the pollutant.

So again, looking at this mass balance model,

again our flow in is going to be flow out.

So we’ve, we’ve settled that.

So our mass going into our control volume is going to be

the flow times the concentration coming in plus M, our mass emissions rate.

So we have lead in the background that’s coming into our control volume

and we have a lead emission rate within that control volume.

Our mass leaving that control volume will be the flow times

the concentration going out at that border plus the depositional velocity times

the cross-sectional area of that city times the concentration.

So again, we said this was going to be a steady state problem,

so mass in is going to equal mass out.

So we can substitute a few terms.

Rather than Q we can use the cross-sectional velocity or

the wind velocity times the cross-sectional area of the Leadville,

substituting that for Q and so again we set up terms mass in,

concentration plus the emission rate, mass out,

flow times concentration out plus the depositional rate.

So again, lead is not being transformed.

It’s either physically coming in or out of the system.

Okay, so we can set up our mass balance equation

for the pollutant that the mass in is going to be the mass out.

We can rearrange terms and solve for concentration out.

We didn’t carry all the units,

but at home make sure that you double check that you carry the units through

and that they’ve been cancelled out properly.

And we get a concentration out at .55 mcg/m3.

Okay, so very similar to the problems we solved before.

Steady state conservative pollutant so we can set the mass in equal to the mass out.

The only thing that’s a little different here is

you have to make sure you understand how we set up the control volume

and that we are given a mass emission rate and lead deposition rate.

So those are again in and out of the system.

Okay, any questions on setting up a mass balance for air?

It’s a control volume like anything else.

We just have to make sure that when you do these calculations

that you carry the units properly.

You may be given units to convert in an exam problem.

So you have to understand how to do that. Okay.

So we if we look at the way our, our mass balance equation

is set up in terms of Cout, the question is how can we lower that Cout?

You know if we want to reduce lead emissions there are a couple obvious answers.

I mean we certainly can reduce the mass emissions rate.

You know if we reduce M, then Cout is going to be reduced.

The other opportunity here is if we have a very low, low background.

If we increase Z, the mixing height or the wind velocity,

it will also reduce the concentration going out.

So you know we don’t have control over these issues,

but certainly we can control the emissions rate for lead.

So now let’s throw another curve ball into the problem.

Let’s say that we now want to find out how long it would take

to reduce our air pollution, our Cout,

if we had a sudden change in emission rate or wind speed.

Let’s say that again we, we found some mechanism to reduce the lead emissions.

Well again now this, is this a steady state or a non-steady state problem?

Steady state, raise your hands.

Uh, no. You’re going to have problems in a couple weeks.

We’re changing, we’re changing the temporal variable here.

We’re looking at a change in time to see a change in, in concentration.

So dmdt is not 0. That’s what we’re looking for.

So we have to go into our, again you,

you’d be given this solve solution for the differential equation.

But the first thing when you hit these problems,

you have to ask yourself is it steady state or not.

So we’re looking at a change in concentration over time,

so it’s going to be non-steady state.

Conservative or non-conservative pollutant?

Conservative?

I know it’s the end of the day on Thursday.

But this will wake you up.

Conservative, yes it is a conservative pollutant.

We haven’t changed anything with respect to lead.

No reactions. Those conditions are the same.

So again, when you look at these problems

you’re going to have to ask yourself is this a steady state problem or not.

Are we looking at a change, dcdt, or not?

Okay. So that’s a way to solve our mass balance equations.

Now we’re going to start talking a little bit about chemistry.

So what is smog? And of course it comes from the term smoke and fog.

In your readings, in that extra handout that you received,

there’s an introduction to the reading that was in the course pack

and it talked about air pollution episodes in London.

And we talked about Donora in class a few, a few lectures ago.

So in those cases, you had water droplets that had both sulfur dioxide

and particulate matter in them,

and that smog again was formed primarily from the particulate matter and smoke.

And the SO2 that absorbed that can cause in particulate matter causes

some serious health effects. Now that does not require any sunlight to form.

That’s, that’s a straight physical phenomenon and the cause of smog

that we have seen in urban environments and the cause of smog that we see in Beijing.

So here’s some photos of some traditional smog scenarios.

It can often occur in the winter when temperature inversions can trap pollutants

and we form these water droplets with sulfur dioxide and particulate matter.

Again, this does not require sunlight.

So, and these are pretty, pretty gnarly conditions.

I don’t know if you any of you’ve experienced these in a heavy urban environment,

but they cause respiratory distress and, and other health effects.

So now we’ll move into photochemical smog and

the one that’s become a much larger issue of concern.

So what can cause smog?

Well again we have these fine particles that are discharged in power plants primarily.

If we are using diesel fuels,

those can also have particulate matter in their discharge pipes.

So again, these fine particles can blend with the water dot,

droplets in fog and cause this smog, which can be deposited.

What we’re going to be looking at further will be photochemical smog,

which is formed primarily due to NOx, meaning NO or NO2.

I’ll use that as an abbreviation, but NOx, nitrogen oxides,

that are discharged along with volatile organic compounds in a variety sources.

Those are discharged from factories.

You know if you have a boiler that is generating heat,

that boiler will discharge NOx and some volatile organic compounds.

Of course at a refinery or other chemical plant as a chemical process,

you can have NOx and, and VOCs discharged.

Vehicles will discharge NO as the primary pollutant and VOCs from the tailpipes.

So you have a number of sources of NOx and VOCs that

I’ll show you can form ozone in the troposphere.

Again this is the bad ozone.

You may recall that the stratospheric ozone,

that ozone blocks some very powerful UV radiation and

so we want to keep that ozone in the atmosphere.

This, this ozone that we’re generating as you’ll see shortly has a number

of health effects and other chemical effects that are, that are not good.

So what’s photochemical smog?

Well we generally have to look at the impact of volatile organic compounds and

NOx in a photochemical reaction that’s going to generate smog.

Now in this definition we’re calling smog essentially the photochemical oxidants primarily ozone.

So this is in again the, these droplets that will contain sulfur dioxide

and particulate matter, we’re talking primarily about production of ozone.

One little number to keep in the back of your mind is that

the National Ambient Air Quality Standard for ozone is 120 ppb,

and so this is the trip for, for other activities.

So what’s the big deal with ozone?

Well it’s an irritant.

Again, it’s an oxidant and so it can irritate your respiratory system.

It can be an eye irritant.

So it’s again something that you don’t want to be around,

especially if you’re sensitive populations.

But there are other effects according to, other effects that follow

from higher ozone levels.

Ozone actually causes 90% of the damage to agriculture.

It can impact tree foliage and can stunt growth.

Crop damage due to ozone may exceed $2 to $3 billion a year,

which is again a significant part of our agricultural productivity, 2% to 3%.

Now the benefit is that again a lot of these agricultural areas

are outside of areas with significant ozone production,

but ozone is transported in ambient air.

So it, it has an impact on our, on our agriculture.

Well tire life, I mean all of us drive vehicles,

but tires breakdown these days not so much due to typical tread wear,

but also by sidewalls deteriorating due to the presence of ozone.

Again ozone is an oxidant.

So it’s going to breakdown rubber and it will have an impact on your tires.

So other evidence of smog damage include fading and cracking of paints

and some accelerated metal corrosion.

So we talked a bit about the chemicals that are responsible for smog production.

When we combust air, you know we use air in to, for combustion,

it’s a mix of primarily nitrogen and oxygen.

Well that combustion will cause the formation of NO, which is our primary pollutant.

But the NO can combine further with oxygen to cause NOx or NO2.

So this NOx is what will react then with sunlight to, to form this oxygen atom.

And this oxygen atom is highly reactive and that will combine with oxygen to form ozone.

So our combustion doesn’t directly form ozone,

but the combustion product, NOx, will again react with sunlight to form this oxygen atom,

which will then combine with oxygen to form ozone.

So ozone doesn’t keep building and building and building.

Ozone is then degraded by NO.

It will react with NO to form NO2 and oxygen.

So you know the, the concentration of ozone in the troposphere

is a balance between ozone production of course and ozone degradation.

So looking at this in kind of a process model.

We have emissions, again whether this comes from a boiler, from a power plant,

from a vehicle, we’re going to produce NO, which will again form to NO2 formation,

which can again cycle back to NO.

That NO2 generates free oxygen, which will react with oxygen,

molecular oxygen to form ozone. So this reaction does require sunlight.

So you’ll see some temporal variation of course, in the production of ozone,

you need sunlight to get this oxygen atom produced to form ozone.

So you see again we have a balance.

The nitrogen dioxide produces free oxygen, the oxygen atom which produces ozone,

and yet we have NO that’s going to destroy ozone,

and that balance is going to determine the ambient ozone level.

So we if we look at equilibrium ozone concentrations

as a function of the initial NOx concentration,

everything is going to be dependent on the ratio of NO2 to NO.

So we can calculate, we can use equilibrium concentrations to generate this,

our ozone concentration as a function of these reaction rate constants,

which will be dependent on sunlight, and our NO2 and NO concentrations.

So if we look at the NO2 to NO ratio at .2, at a ratio of .2,

that equilibrium ozone concentration will be pretty low.

If we start, if we raise that ratio to .6,

our equilibrium ozone concentration will still be low,

but it will still be increasing of course.

But as we get to a ratio of 1,

then our equilibrium concentration starts to produce ozone

and it’s greater than the rate that ozone is, is degraded.

So this equilibrium ratio of NO2 to NO is the,

you know the tripping factor that’ll drive whether

we’re producing ozone or whether we’re degrading it.

So how do VOCs fit in the mix?

Well the VOCs don’t generate ozone directly.

But again we can have production of this molecular,

of this atomic oxygen with water to form these two hydroxyl radicals,

and those hydroxyl radicals are very reactive.

Those radicals will react with a hydrocarbon.

In this case, you know we’re calling R, our hydrocarbon chain,

this can be methane, ethane, you know a larger hydrocarbon chain.

R is just going to be our symbol for those organics.

So it will react with that hydroxyl radical and nitrous oxide and oxygen

to form more of this NO2, as well as the reactive radical

that will continue to form some NO2, and this is kind of an ongoing chain.

So this, these hydrocarbons will generate these free radicals

and kind of fuel the production of NO2.

So what happens is we wind up essentially using up the NO that would typically

be used to degrade the ozone, we’re going to form NO2,

which will thereby form more ozone.

So this is a kind of a, a positive feedback mechanism

where we generate more ozone then is destroyed.

So in Los Angeles we’ve got some geographic factors

that can cause for these buildups of ozone.

The first is we can have temperature inversions.

So essentially our mixing height within the urban area is small.

It essentially puts a lid on our airshed keeping the, essentially reducing

that control volume, so increasing our concentration of ozone.

If we have low wind speeds, again we’ll have fewer air exchanges

and that will also increase the concentration of ozone within that control volume.

Sunlight will drive production of this atomic oxygen,

which will again react with molecular oxygen to form ozone.

And vehicles will produce NO and then NO2, and of course VOCs

that will fuel this free radical production to generate more NO2.

So if we look at the buildup of NO, NO2, and ozone in L.A.,

this is a fairly typical sort of production curve of nitrogen dioxide and,

and nitric oxide in the atmosphere.

Remember that 122 ppb is our maximum daily one-hour average.

So this would be sort of the cut-off for the National Air Ambient Quality Standard.

So typically we can hit that or come close to hitting

that in L.A. in the late morning.

You know we’ve got commuter traffic.

We’re generating a lot of VOCs.

These reactions occur fairly quickly,

so we’ll get an increase production of, of NO2

because of the emissions out of the tailpipe,

as well as the hydroxyl radical formations

from the hydrocarbons that are discharged to get a peak

of ozone formation in the late morning.

If you look in your readings that are some additional figures

that show ozone concentration profiles over time.

There was a figure in the reading from Massachusetts

that showed this sort of morning peak due to commuting and

then a later afternoon/early evening peak because of transportation of,

or migration of ozone into that urban airshed at a later time

from another up gradient site.

So ozone in the local atmosphere will be dependent

not only on the sunlight and wind speeds and temperatures

and that big urban commute in the morning,

but you can also have ozone then transported into your urban airshed. Go ahead.

[Student Inaudible]

Probably because the sunlight goes down.

We don’t have as intense of radiation at that time,

which again drives the formation of, of the NO2 and those radicals.

So it’s fairly typical to see this curve drop off later in the day,

because just the sunlight isn’t strong.

Well here’s some photographs of Santiago, Chile.

Fairly similar scenario with L.A. in that you’ve got an urban population in a valley,

and so late morning things are actually not looking too bad,

but by late afternoon with the buildup of vehicle traffic and the sunlight,

we’ve got smog formation that really impacts visibility.

Okay. So we have talked a bit about our box model for urban ozone formation.

Again, the, the parameters to remember that are important

will be the wind velocity, the emission rate, the deposition velocity,

and of course the geometry that you need to set up your control volume.

So you know and we, we can set up these mass balance equations,

that emission rate is going to be a really important number.

Our deposition rate, which is often quite unknown,

is going to be an important number,

and again whether we have any reaction of the material.

Now we can look at box models for ozone including VOCs inputs

and NOx inputs and that’s of course well beyond the,

the scope of this course, but we can come up with models

that will predict ozone formation based on an initial VOC concentration.

So remember when VOC levels are high there’s,

there’s still plenty to convert NO to NO2.

But we have to remember that, that NO destroys ozone, but NO2 creates it.

So this graph just looks at total carbon in an atmosphere and

we certainly see the increase of ozone production with an increase in VOCs.

Now looking at the magnitude of these things,

you know when you’re already out in the kind of plateau area of this curve,

you can look that you know if you reduce VOCs by 20%,

you’re just not going to have much of an impact on ozone production.

You’re really in an insensitive part of the model to VOC production.

You’ve really got to get down to some much lower concentrations of VOCs.

In this case, you know we went from an 80% reduction in VOCs

and that had a 30% impact on ozone production.

So again these models are, are dependent on VOCs,

but once we get out into these plateau areas

it’s going to be fairly insensitive to any reductions in VOCs.

So we look at ozone creation as a function of the NOx concentration.

We have our initial NOx concentrations and our ozone production actually peaks,

and then it starts to drop off. So why does that happen?

I mean you would think intuitively well if I decrease NOx

then I should decrease ozone.

But remember that when we produce NO2 we’ll also produce some NO at some point

and so as we start to produce NO, that’s going to destroy ozone

and it’s going to decrease the ozone concentrations in our model.

So in this case, if our ozone, if our NOx concentrations are high,

let’s say in that 122 ppb range,

you may think well let’s reduce NOx to reduce our ozone,

but actually according to that equilibrium model,

we’ll actually increase the ozone concentrations because we aren’t producing as much NO,

which degrades the ozone.

So you’ve really got see where you are in the model.

You can’t automatically assume that okay I’m going to reduce VOC concentrations,

I’ll get an ozone reduction, or I’m going to reduce NOx concentrations,

I’ll get an ozone reduction.

It really depends on the actual concentrations and where you are in the model.

Okay. We’re going to look, kind of finish up our lecture today

looking at some isopleth diagrams.

It really can provide a roadmap as to how to deal with elevated ozone concentrations.

So we can use these models to essentially come up with a contour map for ozone.

This is really what this is.

So we’ve plotted the ozone concentrations as a function of VOCs and NOx.

So of course as you know we go and increase VOCs and increase NOx concentrations,

our, essentially our isocontour lines for ozone concentrations here will increase.

So using those model calculations for, you can,

if you have a given concentration of VOC and a given concentration of NOx,

you can predict your result in ozone concentration to see where you are on this map.

So if you for example at .6 ppm of VOCs and about 120 ppb of NOx,

you’d expect an ozone concentration of 240 ppb.

So the, so any, any questions about what we’re looking at on these isopleth figures?

So essentially we’re using our equilibrium model.

We’re inputting a given concentration for VOCs, a given concentration for NOx,

and it’s going to tell us really where we are in terms of the ozone concentration.

So we can use these models to try to decide what we’re going to do if,

you know how to address a given ozone concentration.

So in this model we’ve identified what’s called a ridge line and

it really divides where we are in ozone concentrations with respect to NOx versus VOCs.

So if we are looking at this part of these isopleth diagrams,

we’re essentially NOx limited,

which means that the ozone production is really driven by the NOx concentration.

We can vary VOCs all we want, but we’re not going to impact the VOC concentration at all.

Similarly, where in this part of the isopleth diagram, we’re in a VOC limited regime.

So we can change the NOX concentration really all we want,

but, we can kind of go up and down the NOx curve, the NOx concentration,

but we’re not going to change the ozone concentration very much.

We again this is based on the interplay between VOCs and NOx

and their associated rate constants.

Now we’ve got a couple of different ways

to look at ozone concentrations based on, on VOCs.

One method is to use a carbon mass approach.

So for example here, we’re looking at all VOCs essentially as if they were equal.

So compounds with larger masses would have a larger impact on the ozone production.

And, and that’s not a particularly accurate model.

We have another model where we can look at reactive organic gases.

So rather than looking at VOCs as a bulk,

we can look at specific reactive gases and that’s a,

that’s a little more accurate model.

We can look at overall chemical reactivity and

I’ll show you a table that looks at that,

and that’s again a pretty good representation.

But the best methodology that’s out there now looks at maximum incremental reactivity,

and what that, what this ratio is,

is really looks at the amount of ozone produced for a given change in VOC concentrations.

Not all VOCs are created like, alike.

Some will be responsible for a greater ozone production than others,

and so the, we have an opportunity to look at that

through this maximum incremental reactivity value.

The California Air Quality Board essentially uses these values

to look at emissions of specific VOCs.

So again not all VOCs are, are alike.

So by using these, these maximum incremental reactivity numbers

we can get a much more accurate feel for ozone production for a given VOC emission.

Any questions about these approaches?

Okay. So again there were some problems in your reading

that you should take a look at because you were given a set of VOC, NOx,

and ozone concentrations and you have to decide really

where you were on the curve and whether changing the NOx or

changing the VOCs would have an impact on ozone.

So hint-hint for the exam, make sure you understand how to use these diagrams.

All right.

This is a table that looks at a whole series of VOCs

and compares reactivity constants.

In other words, what kind of hydroxyl radical production rate could we have

versus this maximum incremental reactivity number,

which is again a ratio of ozone formed per, per gram of VOC emitted.

And the interesting thing about this table is that these data don’t always match up.

So the hydroxyl radical formation isn’t necessarily parallel

to the maximum incremental reactivity.

Now the numbers that we’re most concerned about with the MIR

are values that are generally 5 and above.

So if we have compounds like ethane, n-octane, propene, trimethyl benzene,

you know those are going to be responsible

for more significant ozone production than compounds like methane or, or.