Solar Concentrators under Clouds

Quite often we see suggestions for solar energy systems that rely on mirrors or lenses to collect and concentrate insolation. Solar-thermal devices need the higher irradiance to achieve the temperatures that make thermal cycles practical. Some photovoltaic devices, like multi-junction cells, also need higher irradiance to reach their full potential. Unfortunately, there is one giant drawback to using optics to focus sunlight: it doesn't work when it's cloudy.

When the sky is clear, light from the sun arrives in parallel rays that can be easily focused. For an example, I modeled a 1 m diameter parabolic mirror receiving 785.4 W of power (equivalent to 1000 W/m^2). I put a 10 cm diameter collector near the focus to collect the light.


Figure 1: Ray trace of a parabolic concentrator
A detector on the end of my collector can record the irradiance map.

Figure 2: Irradiance map under direct sunlight
As you can see, the parabolic concentrator creates a nice, even irradiance on the collector. Most of the energy incident on the concentrator ends up on the collector, except for the occlusion in the center caused by the collector itself. Some 99.3 % of the incoming light ends up on the collector surface. In the real world, due to Rayleigh scattering off the atmosphere, performance is not quite this good.

However, when the sky clouds over, the amount of light scattered by the atmosphere increases enormously and the light that hits the Earth's surface is diffuse rather than specular. We all observe this through the obvious fact that there are no shadows when it's overcast. The impact on the performance of optics is dramatic. I won't bother showing a ray trace, since it looks like garbage, but when I diffuse the incoming light the concentrator performance drops by two orders of magnitude.

Figure 3: Irradiance map under indirect sunlight
The total irradiance on the detector has dropped from 780 W to 7.6 W. Obviously, this is no longer sufficient to drive a Stirling engine.

The local climate and in particular the clearness index is critical to determine if optics can be used to concentrate solar energy. The clearness index is the ratio of extra-atmospheric insolation to insolation measured at the Earth's surface. The ideal ratio is 1.0. Clearness index data (and lots of other solar relevent information) can be found at NASA's Surface Meteorology and Solar Energy website:

http://www.eso.org/gen-fac/pubs/astclim/espas/world/ION/ion-pwv.html

Some people have been working on non-focusing optics to achieve better economics. They often use thin refractive lenses to focus light to a slightly smaller blob. Generally these are designed to increase irradiance by a factor of four or so, not tens. The goal with non-focusing optics is to use a cheap concentrator to increase the irradiance on an expensive collector (like a photovoltaic cell).

For the curious, I used the optics modeling package Zemax to produce these results.

Plug-in Hybrid Ancillary Services, Take 2

Engineer-Poet provided links to AC Propulsion White Papers on the idea of using a vehicle's battery capacity to perform ancillary services. Alas, it appears I have not yet had an original idea in the field of entropy production.

The top paper is the one to read, A Vehicle-to-Grid Demonstration Project: Grid Regulation Ancillary Service with a Battery Electric Vehicle. It seems they actually put together an electric vehicle and simulated it operating as a voltage regulator. The paper is remarkable numbers free for a real-world system. Since they are talking about an all-electric vehicle, the electricity numbers are way higher than what we need for a plug-in hybrid. In particular, they need special, high power throughput connections to handle the load. I think most people will agree, the all-electric is dead in the water, since it will never be as flexible as a hybrid car.

I did find some issues that I take umbrage with. For one, following the end of driving in the morning, they recharge immediately to about 80 % capacity. I think this is largely impossible from a utility point of view (and a huge strike against the morbid electric concept). The peaking power necessary to simultaneously recharge millions of vehicles at the end of rush hour would be too expensive to imagine. For the plug-in hybrid, it is much more reasonable to trickle recharge slowly over several hours. After all, if you decide to take the car out for lunch, you aren't going to be in danger of running out of juice. You can also drive through more than one state/province per day.

The AC Propulsion paper does explicitly state that the value majority of services are provided while the car is connected at home, relative to the workplace. In a truly deregulated market, power would be cheapest to buy during the dead of night. I.e. the plug-in hybrid can peak shave.

The biggest philosophical difference between my proposal and AC Propulsion's is that they assume their electric can act as a sink and a source. I think the hybrid should be a pure sink. My idea is flat out simpler (and hence better). When power can only travel in one direction you avoid issues like net metering and up-voltage transforming. This does not impact the ability of thousands of plug-ins to do up and down regulation. Let me explain.

When recharging the plug-in, you are trickle charging it at some slow rate. If you need to regulate up, you stop the charging process. If you need to regulate down, you increase the rate of charging. The power flowing into the vehicle can be between zero and the maximum rate. Normally it will be at some median value to minimize the peak power demand on the utility and maximize the value of regulation service that it can provide.

The sort of flexible, predictable demand management provided by the grid-integrated plug-in hybrid is more valuable than supply side ancillary services. Imagine the ocean floor, with all its ridges and shelves. Now imagine the plug-in hybrid demand is the water. Notice how everything has become oh so much flatter. This metaphor essentially explains how flexible demand can fill in all the vibrations in the supply and demand for electrical power.

One concept they do introduce that is helpful is the idea of an aggregator in a deregulated market. The aggregator is a business entity that contracts out your vehicle's ancillary capacity and sells the aggregate, decentralized services of thousands of vehicles on the ancillary market. The aggregator basically acts as the middle-man, insulating the consumer from the complexities of the power market.

Imagine this scenario as an aggregator: an organization leases out the ground levels of all the parkades around your office. All the parking spaces on the ground floor are labeled in yellow 'plug-in ONLY'. You hop out of your car, plug it into the provided inverter, and go grab a coffee with the time you've saved hunting for a parking space. When it's time to go home, you unplug your car and drive out, laughing at the line-up of gasoline powered cars waiting to pay the attendant. The aggregator basically barters you a ground-level parking space. In return, you let them use the batteries on your car to regulate the grid. It is a fine example of creative marketing -- you get quality of life back by buying a hybrid. That's worth more than a few pennies on the kilowatt-hour.

In a deregulated market the aggregator might be able to function without much extra regulatory support. However, most power markets in North America are not deregulated, and legislation would still be needed in those areas.

Coal Gasification Acoming

Canada is rapidly consuming all of its conventional natural gas reserves. Consumption is profligate, especially South of the border. In the USA, supply has not kept pace with demand, due to underinvestment and rampant NIMBYS (Not In My BackYard Syndrome). As a result, the price of natural gas has skyrocketed to around $9.00 / Mbtu. Prices in other parts of the world, such as South America, are almost an order of magnitude less.

Canada is expected to exhaust its conventional natural gas deposits in about 3-4 years. New sources, such as coal-bed methane, will take over the slack, but there is serious danger of a supply disruption. That risk is forcing a lot of chemical industry corporations in Alberta to take a long look at coal gasification to insure they have a stable supply of natural gas. I feel there should be a nuclear steam plant supplying all the bitumen refineries in Fort McMurray, but I don't really have a lot of political influence with King Ralph.

Another possible fix is the construction of a Liquified Natural Gas (LNG) terminal in Prince Rupert. The federal and provincial governments both recently promised money to expand the port, and it is the closest West Coast port to Edmonton (and Alberta's chemical industries). Excess capacity could be shipped down South. Shipping liquified natural gas from South America, to Prince Rupert, and then down back to the Northwest USA may seem absurd. That's generally where NIMBYS gets you, however.

The Feebate

Now, for part two of encouraging the introduction of the (plug-in) hybrid:

The best way I know of to encourage more efficient automobile buying habits is called a feebate. A feebate is a zero-sum tax-slash-rebate on motor vehicle fuel economy. Car buyers with fuel economy above the mean get a cash rebate, while gas guzzlers must pay an additional tax. The government simply acts as the creditor in the exchange -- it gains no net tax revenue. E.g. it is a very socialist concept. The nature of the feebate helps offset the higher capital investment of a hybrid vehicle. Increased mileage is, of course, a built-in reduction in operating costs.

There have been a number of academic papers on the feebate concept. The most recent one, which I will regurgitate here, is

DL Greene et al., Feebates, rebates and gas-guzzler taxes: a study of incentives for increased fuel economy, Energy Policy 33 (2005), pp. 767-775.

The authors examined a number of feebate programs, at a $500 and $1000 per 0.01 bushels per rod rate (or that might be gallons per mile). The impact of the program on average mileage is heavily determined by how far-sighted the consumer is. Since the consumer is myopic with a 4.67x diopter, most of the analysis is based off a 3-year amortization period, rather than the assumed 14-year lifespan of the vehicle. The last key point, is that they aren't using an absolute zero-sum feebate -- there is a very small ding to government finances.

The results?

1. $500 per 0.01 GPM: a 12.5 % gain in car mileage and a 25.1 % gain in light truck mileage.
2. $1000 per 0.01 GPM: a 24.6 % gain in car mileage and a 40.7 % gain in light truck mileage.
   

They also looked at a pure gas-guzzler tax or a pure rebate. Neither is nearly effective as the combination. The results appear quite significant, especially in terms of the impact on the evil soccer mom's SUV.

One of the conclusions is that buying patterns aren't significantly altered by a feebate. Instead, it mostly acts to facilitate the purchase of more technologically advanced models. Overall unit sales are slightly decreased, but the overall monetary value of car sales increases with the higher capital costs. Of course, this favours Japanese automakers over the comparatively mentally retarded domestic manufacturers.

Of course, any results from a study that tries to mathematically model consumer buying habits should be taken with a grain of salt. Treat these numbers as approximate only. I think a key of any feebate program is a truth in advertising provision. If advertisers are forced to include the feebate in their listed prices I think the impact on vehicle choice would be more pronounced. The study cannot effectively model the state of mind of the consumer. As we saw during the Middle-east Oil Crisis, if North Americans think oil is expensive, they will buy more fuel efficient cars. The actual price of oil is not so important.

As usual, I do not support specific subsidies for hybrids themselves. Otherwise we may see the absurdity of the Hummer Hybrid. Energy Outlook has a good discussion of this general issue here:

http://energyoutlook.blogspot.com/2005/08/specifying-path-vs.html

Tailoring Electricity Demand with the Plug-in Hybrid

The plug-in hybrid is a hybrid electric-gasoline powered automobile that has a significantly large battery capacity. Unlike the normal hybrid it can be plugged into the utility grid, allowing its batteries to be charged up with electricity rather than the gasoline engine. Some people like to call the plug-in hybrid the GO-HEV (Gasoline Optional-Hybrid Electric Vehicle). Personally, I have a pathological hatred of acronyms.

Since most people only commute a short distance every day, the plug-in has the potential to shift the majority of the transportation load off oil and onto electricity. At the same time, the plug-in hybrid retains the range and acceleration of its simpler basic hybrid cousin.

According to the Office of Energy Efficiency (of Canada) the total annual energy consumption of the passenger transportation sector was 1,322.4 PJ in 2003. That number corresponds closely the with the gasoline consumption of the nation. The freight transport sector consumed 945.8 PJ (mostly diesel). The total overall energy consumption was 8,457.3 PJ, so passenger transport consumes about 15.6 % of the total energy production of the country. About half of the cars on the road are driven less than 30 km a day. Current plug-in hybrid technology is easily capable of meeting this demand. Thus, in a country of about 33 million people, the plug-in hybrid is realistically capable of shifting 500 PJ of load from oil to electricity. This works out to about 11.5 kWh per person, per day! That's a lot of electricity demand. I personally only consume about 7 kWh per day, and I live alone, have electric baseboard heat, and an electric range.

The major drawback of the plug-in hybrid is its higher capital cost from the extra battery capacity. The other issue of the plug-in is that it is hardly a better solution if the electricity comes from coal or some other polluting resource.

The solution is two-fold. The first solution is to give the utility companies charging control over the plug-ins. This necessitates not only a 220 V wall plug, but an internet connection for a plug-in hybrid. The other side of the plug-in hybrid coin is a feebate program to reduce the associated capital costs. I will have to talk about that in another post.

The advantage here is that it creates a large, flexible demand for the utilities to fill (obviously the utility will have to guarantee a given level of charge in the morning). Why does this matter? Let's take a look at Texas' daily ramping power demand:

http://currentenergy.lbl.gov/tx/index.php


We can see that the demand varies from about 30,000 MW in the early morning to a peak of 50,000 MW at mid-day. Texas' grid is almost completely independent of the rest of the USA. There's probably less than 1000 MW of transmission connectivity between Texas and the outside world. That's a big shift, as it means during the course of a day almost 70 % more capacity has to be warmed up and brought on-line. Generally speaking, this doesn't happen. Power produced in the morning is often simply wasted (known as power shedding). Giving the utilities control over charging hybrids will give them the ability to fill in that valley, operating more efficiently by smoothing the demand variance, or having the demand follow supply rather than the other way around. Of further benefit, it will allow much more reliable forecasting, so power will not be wasted by overestimating demand.

Just how much flexible demand can the plug-in hybrid create? Consider that I claimed plug-in hybrids could supplant 500 PJ per year for Canada, and Canada consumed about 1900 PJ of electricity in 2003. Then slightly over 20 % of total electricity demand could become 'flexible' through the wide-spread introduction of the plug-in.

Of course, plug-in owners need to be compensated for providing this service to the utilities. Collectively, transmission services such as load-following, voltage-regulation, spinning reserve, etc. are known as ancillary services in the grid world. With controllable demand, the plug-in can take over many of these functions when acting as a demand sink. How valuable are ancillary services? Typically they will run from $0.01 - 0.03 / kWh, a significant chunk of the cost of electricity. If plug-in owners are paid for the ancillary services they are de facto selling, and their rates are reduced appropriately, they could see a very significant reduction in the cost of running their vehicle.

This flexible demand is valuable now, with our current energy grid. If we introduce more and more intermittent sources (i.e. wind, solar, tidal, wave), that flexibility will become even more valuable. Hence renewables and the plug-in hybrid are complementary technologies.

This idea of course, will go nowhere without government intervention. As hybrid vehicles become more common, and the introduction of the plug-in appears on the horizon, it would be prudent for governments to establish rules reguarding their use and interaction with the grid. Otherwise we risk extra-strain on the grid, as demand skyrockets whenever rush hour ends and millions of plug-ins start eating electrons.

PV Shortage

The NY Times is carrying a story on the tight supply of PV cells in the world right now (registration required):

Shortages Stifle a Boom Time for the Solar Industry

The good news is there's a large increase in factory production capacity coming on-line in 2006. The bad news is that polycrystalline Si-cells are made from microprocessor industry waste Silicon. Once the PV industry runs out of waste Silicon to buy on the cheap and has to zone refine and grow its own crystal, the cost will increase permanently.

This is a good example of Germany "picking the winner" technology, even though it might not be economical. The Danes have 'picked' wind power for example. If they didn't have Norway's hydropower to tap into they would be screwed right now; the rest of their electricity production is cogeneration coal which can't load-follow worth a damn.

Government and politicians picking environmental power technologies is a good way to give environmentalism a black eye over creditability. In the future, this PV craze may hurt the economy of California and Germany through a poor return on investment. Photovoltaic could suffer the same credibility failure of nuclear that was going to be "too cheap to meter."

Subsidies should be directly aimed at CO2 and other pollutant emissions, not particular technologies. Let engineering and the economy sort out the winners and the losers on their own.

EDIT (August 8th):

I don't think I made my point very clear in my original post.

What I want to say is that these cost inefficient silicon solar cells have no long term potential for on-grid applications. They are way too expensive. Only novel photovoltaic and solar thermal projects have the potential to produce economical electricity.

These subsidies do nothing to encourage corporations to invest in solar power innovation. Instead, they only encourage companies like BP Solar and Sharp to over invest in Si-photovoltaic production capacity. There is no long term value in this investment.

If you want an example of how screwy government subsidies can stifle innovation look at how Boeing and Lockmart operate in the US aerospace industry.

Once publicly traded corporations start feeding at the public trough they are loathe to stop.

4 Days in Hawaii

I just spent 4 days of my life attending the Microscopy and Microanalysis 2005 conference in Waikiki. I think Waikiki in Hawaiian means 'tourist trap hell'. Contrary to expectations, Waikiki beach is not populated by Maxim girls.

I was fortunate enough to win a student travel award from the Microscopy Society of America to go there and do a platform presentation. I even got a plaque, with gold leaf embossment. I got really excited as I imagined the surface plasmons of the glided plaque coupling with the fluorescent lighting to create that glorious and lustrous shine. The back side has a bevy of mounting options, so I can display it to my office mates at school. (This post has long been tongue in check if you haven't noticed.)

I learned a number of things by attending the conference.

1. Electron microscopes are phallus shaped -- everyone likes to compare.
2. Everyone uses their 'unit' to look at god, err... gold even. The more money you have, the better you can see god.
3. If you don't have the latest-greatest accessory for your unit, you can't do good research, unless you are a graduate student.
4. 15 minutes is not enough time to say anything aside from, "clear as mud?"
5. Don't try and do wind sprints after being on an airplane or bus for 20 straight hours unless you want to pull both hip flexors.

Sweeping CO2 Under the Rug

An examination of Canada's Carbon Dioxide production suggests Canada is one of the worst carbon polluters in the world, at 18.9 tonnes per capita.

http://www.eia.doe.gov/emeu/cabs/canada.html

The USA still stands above us at 20.3 tonnes per, but the numbers are surprising because 70 % of Canadian electricity is from hydroelectric and nuclear -- non-CO2 emitting sources. In comparison only about 25 % of US electricity doesn't come from carbon fuels. So where's the extra CO2 coming from?

It's not the transportation sector. The source is, in fact, industry. Canada's industry is one of the most energy intensive in the entire world. This is largely due to the fact that Canada has a heavy resource-based economy. Canadians consume about 23% more energy per person than even our energy hog friends to the South.

Pulp and paper, smelting, steel, cement, fertilizer, and bitumen (tar sand) production compose the bulk of energy consumption by Canada's industry. However, not all industries produce equal CO2 emissions for a given energy input. Alcan's Aluminium consumes large amounts of electricity for electrolysis -- it cannot use heat directly from a carbon fuel, it can only use low entropy electricity. Steel, on the other hand, is dependant on high quality coke for reduction reactions to remove impurities. Most of the other industrial processes require either heat or steam.

The Alberta tar sands are a special case. Large amounts of hot steam are needed to separate out the valuable volatiles from the sticky mess. So how are we getting oil from the tar sands? We burn natural gas of course. This makes tar sand gasoline extremely intensive in C02 terms. One possible solution is to build a nuclear steam plant up in Fort McMurray. However, using nuclear to provide heat and steam only really works in that one place, because all the refineries are pretty much lined up in a row. Saskatchewan's potash plants, in comparison, are not close together.

This leaves us at the possibility of recovering CO2 emissions at the source and pumping it underground. Recovery can be done by either physical means (liquefaction of C02 from the exhaust) or chemical means. A technical discussion of CO2 recovery can be found here:

http://asme.pinetec.com/ijpgc2000/data/html/15056.html

Pumping CO2 underground is a relatively well established technology for oil wells. The Saskatchewan Weyburn field project is representative of the concept. As a side benefit, liquefied C02 is actually a superior agent for forcing the remnants of oil from a nearly exhausted field than water. Hence, there might actually be some real economic value in liquefied CO2.

There has been a recent surge in interest in CO2 sequestration by some of the Kyoto opponents in the world. This is of course, somewhat amusing because Kyoto is a purely carbon metric, and it is difficult to not be somewhat cynical of their motives. The real value of Kyoto is that is creates a carbon pollution trading system. Unfortunately, the voluntary system proposed by the USA is likely to be far less successful than the capitalist Kyoto system.

If we can't put a dollar value on liquefied Carbon Dioxide, industry will continue to deal with it the way they know how: with an exhaust.

Oil to Electricity

It used to be that we had a coal economy. Industry ran on it, trains and ships burnt it to supply transportation.

With the introduction of the automobile, we saw a demand for a portable liquid energy source. That led in turn to a shift in the energy industry away from coal to oil.

Currently we appear to be in the transition from an oil to a natural gas economy. It represents a fairly small economic potential gap to jump. There is a fairly large amount of natural gas available for consumption, especially if you consider gasified coal. Natural gas can be liquefied at about 100 K, which is cryogenic but fairly easily reachable with conventional and established refrigeration technology. Natural gas is already routinely liquefied for transport in ships.

Because I foresee this upcoming shift to natural gas, I would regard myself as a peak oil optimist. The peak in oil production may be here already, but I don't think our economy is about to collapse on itself.

Of course, switching from oil to natural gas does almost nothing for the global warming problem. As a Canadian, global warming is not a major negative for my own self-interest. Countries with vast artic territories like Russia and Canada will both benefit and suffer from global warming. However, the world as a whole will need to try and at least collar this problem.

The only real solution that I can foresee is switching from carbon-based energy to some other means of energy storage. Hydrogen is a straw-man, but I can see electricity becoming our new means of energy trading. Electricity infrastructure, obviously, is already very strongly established. What is patently missing is a means of electricity storage. This is especially necessary to make intermittent renewable sources more economical. To this, I look to the proliferation of (non-hydrogen) electricity storage methods: pump hydro, compressed air/natural gas, flywheels, solid oxide batteries, flow batteries, and our standard electrolytic cells. Plug-in hybrids will allow us to shift much of our transport energy consumption from oil/gas to electricity as well.

For this reason, I think the best step any government could take today to move towards carbon independence is the construction of a large DC transmission grid. In order to operate an electricity economy, we will need to be able to transmit electricity from one side of the country to the other. DC transmission losses are usually on the order of 0.6 % per 100 km. Currently, our grids are not well connected. Some grids, like Texas for example, are practically independent of the rest of the continent.

This is a step that we can take now, without relying on vapourware. We know it will save money. It may end up not being the most efficient investment, but we can be confident it will have a higher rate of return than PV or biofuel subsidies.

Renewable Red Herrings

One of the great dangers environmentalism faces in reaching for technological solutions to existing pollution problems is that they've got it wrong. Let's face it, environmentalists tend not to do the math. When environmentalists overreach or exaggerate, they damage the credibility of the green movement as a whole.

In the current landscape, megatonnes of Carbon Dioxide have become the conscience of environmentalism, with leaders like Tony Blair championing the danger of climate change while the US administration hires Exxon to redact its scientific reports on the subject. One of the possible solutions to the consumption of carbon fuels put forth by environmentalists is the development of renewable energy: solar (specifically photovoltaic), wind, biomass, and reservoir hydro.

Of these technologies, the only proven one is hydro-electric dams... Wind has made strides in installed capacity, but its volatile power output make it a liability to the grid. It is cost effective when the wind is blowing, but the fickleness of mother nature puts an excessive stress on gas-fired load-following plants and voltage regulation. Currently the fate of wind hangs in the balance. With increased decentralization, better coupling to strong load-following power providers like hydro, and the introduction of utility-scale electrify storage like flow batteries, it will be able to provide a significant reduction in CO2 emissions.

That leaves photovoltaic and biomass as underdeployed, but potentially promising solutions. Unfortunately, I believe both of these technologies will be failures over the next 25 years.

Photovoltaic cells are energy intensive to manufacture. Optimistically, the energy return ratio of PV cells is about 20:1. This means that for every Joule used in the manufacturing process, the solar cell will only return twenty times that over the lifetime of the cell. Of course, most cells aren't installed in an ideal clearness index climate, oriented to the sun. Installations on home roofs are usually installed at the wrong angle, and may not be south facing. A ratio of 10:1 is more common in the literature. Compared to wind at 40:1 and hydro at 100:1 , PV is not economically feasible.

Furthermore, all current PV production is done using silicon recycled from the microchip industry. This reduces the material cost of photovoltaic manufacturing. The current prices of solar electric cells, expensive as they are, are not sustainable if demand increases substantially.

While decentralization has its advantages, I think solar has picked the wrong method for gathering power. Photovoltaic, realistically, should stay off the grid for the time being. Centralized concentrated solar-thermal collectors seem to have much greater potential for providing peaking power to the grid. However, it should be noted, concentrated solar performs poorly in the diffuse lighting conditions of overcast skies. Concentrated solar, like wind, often requires the installation of expensive transmission lines to bring power from the boondocks to urban areas.

Biomass is a net zero carbon emitter, since plants consume CO2 from the atmosphere and then we later burn them to liberate energy. Compared to other renewables, biomass is storable and portable. As such, it can be used for transportation.

Unfortunately, photosynthesis is an inefficient solar cell. The amount of arable land necessary requires a lot of energy to fertilize, plant, irrigate, and harvest. The energy return of ethanol is terrible, typically around 1.1:1 if it is even positive. Turning over huge tracts of land for biodiesel and ethanol production will have to come out of land being used for food production. While the world is currently feeding itself, we can't do both food and biomass production at the same time.

PV subsidies in Germany and California, along with ethanol and biodiesel subsidies in America and Europe are essentially cases of government picking the winners. Given the historical success rate of government investment in technologies.