In green tech, there are a few markets --solarand efficiency come to mind -- where startups stand a halfway decent chance of becoming a going concern or a ripe acquisition target. Efficiency revolves around hardware and software, while solar seems to give birth to new sub-niches all the time.  

You’re not going to start a crystalline silicon solar module company in the U.S. today, but a company that can produce thin, inexpensive crystalline wafers for Chinese module makers could enjoy a bright future.

Power electronics -- high-demand, novel concepts, and a lot of established chipmakers have been dozing on this issue -- could become a robust sub-market for acquisitions.

And then there are the industries where doom lurks around every corner. Consider biofuels. In 2005, biofuels looked easy. Peak oil loomed, feedstocks like wood chips, algae, and corn stover had little economic value. Plus, the people making money in this business lived in Houston: if they were so smart, how come they live there, VCs asked themselves.

The answer came hundreds of millions of dollars later. Growing algae in ponds might be cheap, but separating the algae from water has been an unsolved technical issue since the late '70s. Microbes that convert forest waste to alcohol can die at low levels of fuel production. Distribution is another hurdle: Don Paul, the former CTO of Chevron, once estimated that it takes about $3 billion and 15 years to get a new fuel from the lab to the market.

The rising price of oil and a growing interest among large corporations to stabilize their commodity expenses has helped revive the market. Amyris, Solazyme, and Gevo have all recently successfully held IPOs. All three companies, however, remain in the early stages of development. All three have targeted specialty chemicals -- which can fetch higher prices -- instead of fuel.

Home networking? Unlike commercial building management, the gains from home automation are small on a monthly basis compared to the cost of a system, so utilities and demand response providers will have to subsidize this. In turn, that means for every company like Tendril or EcoFactor that wins a major contract to blanket a territory, there will be 16 also-rans. Approximately 108 home networking companies have received VC funding. “Not too many of them have made a lot of traction,” said Steve Goldberg of Venrock recently.

But home networking companies have it easy compared to water startups. Electric and gas utilities are zippy Web 2.0-like outfits compared to municipal water districts, which are even more strapped for money than their counterparts are for power. Desalination plants must also pass through the colonoscopy of public land use hearings.

Electric cars? Tesla made it: even if it gets acquired it will likely survive as a brand-within-a-brand. But most other EV startups are either listing (Aptera, Next Vehicles) or planning to release cars at the same time as similar, less expensive models from established car makers.  

Carbon capture? With no price on carbon, carbon dioxide needs to be transformed into a commercially viable product, like liquid fuel or minerals. Unfortunately, CO2 is a relatively inert molecule so the energy inputs can outweigh the benefits.

Hydrogen? Like CO2 conversion with degrees of difficulty added.

Fuel cells? One word: Ballard.

Building materials? Without carbon pricing, green building materials are a luxury. Building is in a slump, too. Serious Materials has yet to build its green drywall factory: instead, it concentrates now on building controls.

Solid state lighting? It will become a $100 billion business with a lot of successful new entrants, but giants like Cree and Osram are solidifying their positions. Startups will need to find ways -- see Intematix -- to thrive in protectable niches.

My personal choice for the most difficult business right now is car charging as a service. Compare the real estate churn of a charging station to a gas station. Consumers stay at gas stations for about ten minutes, spend 50 dollars and quickly leave to allow the next consumer to fill up.

At charging stations, consumers will spend eight hours, colonize one of the few available outlets for an entire day, and spend maybe $2.50.

And that’s when they decide to buy from you. Unlike petroleum, consumers can get electricity at home any time they want. Gas stations are like Starbucks outlets. EV stations are like used record stores where slackers lounge about all day and then complain about the selection. You could see this market easily becoming a huge hardware market, but it would be focused, and conglomerates like GE and Schneider have also begun to market their own charging equipment.

There are only, as far as I can tell, four ways out of the charging quandary. One, let parking companies like Ampco take over EV charging. That would eliminate most of the real estate and overhead costs. Two, get retailers and hotels to buy EV chargers and give away the power as a way to build customer loyalty.

Three, subsidize charging through advertising. Ecotality wants to do this. It might work, but since most customers won’t stay at the pump to see the ad (like the videos now streaming at gas stations), it will take some time to craft the right promotional opportunities.

Four, the Better Place model of combining battery leasing with electricity in subscription fees. Yes, the concept is replete with assumptions, unknowns, high capital costs and infrastructure challenges, but at least you get something -- a battery, a less expensive car, fewer headaches if your first battery wears out quicker than normal -- for your subscription fees. The fact that this makes more sense to me now than in the past is either a sign that the economic and technological circumstances are making the idea more attractive or that I'm inadvertently sniffing paint fumes at my desk.

But what do you think? If you had to start a company today, what’s the one sector you’d avoid like a case of chlamydia?