Tag Archives: water

The chemistry of sewage treatment

6 Replies

The first thing to know about sewage is that it’s mostly water and only about 250 ppm solids. That is, if you boiled down a pot of sewage, only about 1/40 of 1% of it would remain as solids at the bottom of the pot. There would be some dried poop, some bits of lint and soap, the remains of potato peelings… Mostly, the sewage is water, and mostly it would have boiled away. The second thing to know, is that the solids, the bio-solids, are a lot like soil but better: more valuable, brown gold if used right. While our county mostly burns and landfills the solids remnant of our treated sewage, the wiser choice would be to convert it to fertilizer. Here is a comparison between the composition of soil and bio-solids.

The composition of soil and the composition of bio-solid waste. biosolids are like soil, just better.

The composition of soil and the composition of bio-solid waste. biosolids are like soil, just better.

Most of Oakland’s sewage goes to Detroit where they mostly dry and burn it, and land fill the rest. These processes are expensive and engineering- problematic. It takes a lot of energy to dry these solids to the point where they burn (they’re like really wet wood), and even then they don’t burn nicely. As shown above, the biosolids contain lots of sulfur and that makes combustion smelly. They also contain nitrate, and that makes combustion dangerous. It’s sort of like burning natural gun powder.

The preferred solution is partial combustion (oxidation) at room temperature by bacteria followed by conversion to fertilizer. In Detroit we do this first stage of treatment, the slow partial combustion by bacteria. Consider glucose, a typical carbohydrate,

-HCOH- + O2 –> CO2 + H2O.    ∆G°= -114.6 kcal/mol.

The value of ∆G°, is relevant as a determinate of whether the reaction will proceed. A negative value of ∆G°, as above, indicates that the reaction can progress substantially to completion at standard conditions of 25°C and 1 atm pressure. In a sewage plant, many different carbohydrates are treated by many different bacteria (amoebae, paramnesia, and lactobacilli), and the temperature is slightly cooler than room, about 10-15°C, but this value of ∆G° suggests that near total biological oxidation is possible.

The Detroit plant, like most others, do this biological oxidation treatment using either large stirred tanks, of million gallon volume or so, or in flow reactors with a large fraction of cellular-material returning as recycle. Recycle is needed also in the stirred tank process because of the low solid content. The reaction is approximately first order in oxygen, carbohydrate, and bacteria. Thus a 50% cell recycle more or less doubles the speed of the reaction. Air is typically bubbled through the reactor to provide the oxygen, but in Detroit, pure oxygen is used. About half the organic carbon is oxidized and the remainder is sent to a settling pond. The decant (top) water is sent for “polishing” and dumped in the river, while the goop (the bottom) is currently dried for burning or carted off for landfill. The Holly, MI sewage plant uses a heterogeneous reactors for the oxidation: a trickle bed followed by a rotating disk contractor. These have higher bio-content and thus lower area demands and separation costs, but there is a somewhat higher capital cost.

A major component of bio-solids is nitrogen. Much of this in enters the form of urea, NH2-CO-NH2. In an oxidizing environment, bacteria turns the urea and other nitrogen compounds into nitrate. Consider the reaction the presence of washing soda, Na2CO3. The urea is turned into nitrate, a product suitable for gun powder manufacture. The value of ∆G° is negative, and the reaction is highly favorable.

NH2-CO-NH2 + Na2CO3 + 4 O2 –> 2 Na(NO3) + 2 CO2 + 2 H2O.     ∆G° = -177.5 kcal/mol

The mixture of nitrates and dry bio-solids is highly flammable, and there was recently a fire in the Detroit biosolids dryer. If we wished to make fertilizer, we’d probably want to replace the drier with a further stage of bio-treatment. In Wisconsin, and on a smaller scale in Oakland MI, biosolids are treated by higher temperature (thermophilic) bacteria in the absence of air, that is anaerobically. Anaerobic digestion produces hydrogen and methane, and produces highly useful forms of organic carbon.

2 (-HCOH-) –> CO2 + CH4        ∆G° = -33.7 Kcal/mol

3 (-HCOH-) + H2O –> -CH2COOH + CO2 +  2 1/2 H2        ∆G° = -21.9 kcal/mol

In a well-designed plant, the methane is recovered to provide heat to the plant, and sometimes to generate power. In Wisconsin, enough methane is produced to cook the fertilizer to sterilization. The product is called “Milorganite” as much of it comes from Milwaukee and much of the nitrate is bound to organics.

Egg-shaped, anaerobic biosolid digestors.

Egg-shaped, anaerobic biosolid digestors, Singapore.

The hydrogen could be recovered too, but typically reacts further within the anaerobic digester. Typically it will reduce the iron oxide in the biosolids from the brown, ferric form, Fe2O3, to black FeO.  In a reducing atmosphere,

Fe2O3 + H2 –> 2 FeO + H2O.

Fe2O3 is the reason leaves turn brown in the fall and is the reason that most poop is brown. FeO is the reason that composted soil is typically black. You’ll notice that swamps are filled with black goo, that’s because of a lack of oxygen at the bottom. Sulphate and phosphorous can be bound to ferrous iron and this is good for fertilizer. Generally you want the reduction reactions to go no further.

Weir dam on the river dour. Used to manage floods, increase residence time, and oxygenate the flow.

Weir dam on the river Dour in Scotland. Dams of this type increase residence time, and oxygenate the flow. They’re good for fish, pollution, and flooding.

When allowed to continue, the hydrogen produced by anaerobic digestion begins to reduce sulfate to H2S.

NaSO4 + 4.5 H2 –>  NaOH + 3H2O + H2S.

I’m running for Oakland county, MI water commissioner, and one of my aims is to stop wasting our biosolids. Oakland produces nearly 1000,000 pounds of dry biosolids per day. This is either a blessing or a curse depending on how we use it.

Another issue, Oakland county dumps unpasteurized, smelly black goo into Lake St. Clair every other week, whenever it rains more than one inch. I’d like to stop this by separating the storm and “sanitary” sewage. There is a capital cost, but it can save money because we’d no longer have to pay to treat our rainwater at the Detroit sewage plant. To clean the storm runoff, I’d use mini wetlands and weir dams to increase residence time and provide oxygen. Done right, it would look beautiful and would avoid the flash floods. It should also bring natural fish back to the Clinton River.

Robert Buxbaum, May 24 – Sept. 15, 2016 Thermodynamics plays a big role in my posts. You can show that, when the global ∆G is negative, there is an increase in the entropy of the universe.

Weird flow calculation

Here is a nice flow problem, suitable for those planning to take the professional engineers test. The problem concerns weir dams. These are dams with a notch in them, somethings rectangular, as below, but in this case a V-shaped notch. Weir dams with either sort of notch can be used to prevent flooding and improve the water, but they also provide a way to measure the flow of water during a flood. That’s the point of the problem below.

A series of weir dams on Blackman Stream, Maine. These are thick, rectangular weirs.

A series of weir dams with rectangular weirs in Maine.

You’ve got a classic V weir on a dam, but it is not a knife-edge weir, nor is it rectangular or compound as in the picture at right. Instead it is nearly 90°, not very tall, and both the dam and weir have rounded leads. Because the weir is of non-standard shape, thick and rounded, you can not use the flow equation found in standard tables or on the internet. Instead, you decide to use a bucket and stopwatch to determine that the flow during a relatively dry period. You measure 0.8 gal/sec when the water height is 3″ in the weir. During the rain-storm some days later, you measure that there are 12″ of water in the weir. The flow is too great for you to measure with a bucket and stopwatch, but you still want to know what the flow is. Give a good estimate of the flow based on the information you have.

As a hint, notice that the flow in the V weir is self-similar. That is, though you may not know what the pattern of flow will be, you can expect it will be stretched the same for all heights.

As to why anyone would use this type of weir: they are easier to build and maintain than the research-standard, knife edge; they look nicer, and they are sturdier. Here’s my essay in praise of the use of dams. How dams on drains and rivers could help oxygenate the water, and to help increase the retention time to provide for natural bio-remediation.

If you’ve missed the previous problem, here it is: If you have a U-shaped drain or river-bed, and you use a small dam or weir to double the water height, what is the effect on water speed and average retention time. Work it out yourself, or go here to see my solution.

Robert Buxbaum. May 20-Sept 20, 2016. I’m running for drain commissioner. send me your answers to this problem, or money for my campaign, and win a campaign button. Currently, as best I can tell, there are no calibrated weirs or other flow meters on any of the rivers in the county, or on any of the sewers. We need to know because every engineering decision is based on the flow. Another thought: I’d like to separate our combined sewers and daylight some of our hidden drains.

Weir dams to slow the flow and save our lakes

3 Replies

As part of explaining why I want to add weir dams to the Red Run drain, and some other of our Oakland county drains, I posed the following math/ engineering problem: if a weir dam is used to double the depth of water in a drain, show that this increases the residence time by a factor of 2.8 and reduces the flow speed by 1/2.8. Here is my solution.

A series of weir dams on Blackman Stream, Maine. Mine would be about as tall, but somewhat further apart.

A series of weir dams on Blackman Stream, Maine. Mine would be about as tall, but wider and further apart. The dams provide oxygenation and hold back sludge.

Let’s assume the shape of the bottom of the drain is a parabola, e.g. y = x2 , and that the dams are spaced far enough apart that their volume is small compared to the volume of water. We now use integral calculus to calculate how the volume of water per mile, V is affected by water height:  V =2XY- ∫ y dx = 2XY- 2/3 X3 =  4/3 Y√Y. Here, capital Y is the height of water in the drain, and capital X is the horizontal distance of the water edge from the drain centerline. For a parabolic-bottomed drain, if you double the height Y, you increase the volume of water per mile by 2√2. That’s 2.83, or about 2.8 once you assume some volume to the dams.

To find how this affects residence time and velocity, note that the dam does not affect the volumetric flow rate, Q (gallons per hour). If we measure V in gallons per mile of drain, we find that the residence time per mile of drain (hours) is V/Q and that the speed (miles per hour) is Q/V. Increasing V by 2.8 increases the residence time by 2.8 and decreases the speed to 1/2.8 of its former value.

Why is this important? Decreasing the flow speed by even a little decreases the soil erosion by a lot. The hydrodynamic lift pressure on rocks or soil is proportional to flow speed-squared. Also, the more residence time and the more oxygen in the water, the more bio-remediation takes place in the drain. The dams slow the flow and promote oxygenation by the splashing over the weirs. Cells, bugs and fish do the rest; e.g. -HCOH- + O2 –> CO2 + H2O. Without oxygen, the fish die of suffocation, and this is a problem we’re already seeing in Lake St. Clair. Adding a dam saves the fish and turns the run into a living waterway instead of a smelly sewer. Of course, more is needed to take care of really major flood-rains. If all we provide is a weir, the water will rise far over the top, and the run will erode no better (or worse) than it did before. To reduce the speed during those major flood events, I would like to add a low bicycle path and some flood-zone picnic areas: just what you’d see on Michigan State’s campus, by the river.

Dr. Robert E. Buxbaum, May 12, 2016. I’d also like to daylight some rivers, and separate our storm and toilet sewage, but those are longer-term projects. Elect me water commissioner.

A run runs through it

3 Replies

The word ‘run’ appears to be a Michigan dialect for small river. Perhaps Michigan’s most famous run is the Willow run, where the airport is. Currently, almost all of our runs are unrecognizable, they are either trapped in pipes underground, or so dredged out and poisoned that they are more properly called sewers. If I’m elected Oakland county water resources commissioner (drain commissioner) I’d like to free some of these runs, and detoxify them.

These branches of the red run flow beneath the surface of Royal Oak with the main section beneath Vinsetta Blvd.

These branches of the red run flow beneath the surface of Royal Oak with the main section beneath Vinsetta Blvd.

Consider this historical map of Royal Oak. It shows two  river branches, currently under ground. Back in the day, these were known as the north and south branch of the Red run. The south branch is fed by the Washington creek and the small run, now under ground, with the main branch of the run crossing Woodward ave at Catalpa st. These runs only appear above ground in Warren, MI, miles away, as a polluted sewer. But in Royal Oak they should still be clean. If they were partially freed. That is if the channel were exposed to air again to provide small wetlands along the original path — along Vinsetta Blvd, for example. Vinsetta Blvd. already has concrete bridges to show where the run originally ran. The small wetlands would provide habitat for birds and butterflies, and would provide storm relief and some bioremediation as well. After a heavy rain, most of the water would be absorbed into the ground, while the existing pipes carry away the rest.

Robert E. Buxbaum, March 21, 2016

Follow the feces; how to stop the poisoning

4 Replies

In Oakland county, we regularly poison our basements and our lake St Clair beaches with feces — and potentially our water supply too. We have a combined storm and sanitary sewer system that mixes feces-laden sanitary sewage with rainwater, and our pipes are too old and small to handle the amount of storm water from our larger rains. A group called “Save Lake St. Clair” is up in arms but the current commissioner claims the fault is not his. It’s global warming, he says, and the rains are bigger now. Maybe, or maybe the fault is wealth: more and more of the county is covered by asphalt, so less rain water can soak in the ground. Whatever the cause, the Commissioner should deal with it (I’m running for water commissioner, BTW). As the chart of toxic outfalls shows, we’ve had regular toxic sewage discharges into the Red Run basically every other week, with no exceptional rainfalls: only 0.9″ to 1.42″.

Toxic outfalls into lake St Clair, Feb 20 to Mar 20, 2016. There were also two outfalls into the Rouge in this period. These are too many to claim they are once in hundred-year events.

Toxic outfalls into lake St Clair, Feb 20 to Mar 20, 2016. There were also two outfalls into the Rouge in this period. These are too many to claim they are once in hundred-year events.

Because we have a combined system, the liquid level rises in our sewers whenever it rains. When the level is above the level of a basement floor drain, mixed sewage comes up into the basement. A mix of storm water comes up mixed with poop and anything else you and your neighbors flush down. Mixed sewage can come up even if the sewers were separate, but far less often. Currently most of the dry outfall from our old, combined sewers is sent to Detroit’s Waste Water Treatment plant near Zug Island. When there is a heavy rain, the pipe to Zug is overwhelmed. We avoid flooding your basement every other week by diverting as much as we can of the mixed storm water and septic sewage to lake St. Clair. This is poop, barely treated, and the fishermen and environmentalists hate it.

The beaches along Lake St Clair are closed every other week: whenever the pipes to Detroit start getting overwhelmed, whenever there is about 1″ or rain. Worse yet, the sewage is enters the lake just upstream of the water intake on Belle Isle, see map below. Overflow sewage follows the red lines entering the Clinton River through the GW Kuhn — Red Run Drain or through the North Branch off the River. From there it flows out into Lake St. Clair near Selfridge ANG, generally hugging the Michigan shore of the lake, following the light blue line to poison the metro beaches. it enters the water intake for the majority of Oakland County at the Belle Island water intakes, lower left.

Follow the feces to see why our beeches are polluted. It's just plain incompetence.

The storm water plus septic sewage mix is not dumped raw into lake St. Clair, but it’s nearly raw. The only treatment is to be spritzed with bleach in the Red Run Drain. The result is mats of black gunk with floating turds, toilet paper and tampons. This water is filtered before we drink it, and it’s sprayed with more chlorine, but that’s not OK. We can do much better than this. We don’t have to regularly dump poop into the river just upstream of our water intake. I favor a two-prong solution.

The first, quick solution is to have better pumps to send the sewage to Detroit. This is surprisingly expensive since we still have to treat the rain water. Also it doesn’t take care of the biggest rains; there is a limit to what our pipes will handle, but it stops some basement flooding, and it avoids some poisoning of our beaches and drinking water.

This is our combined sewer system showing a tunnel cistern (yuk) and the outflow into the Red Run. We can do better

A combined sewer system showing a tunnel cistern. Outflow goes into the Red Run. We can do better.

A second, longer term solution is to disentangle the septic from the storm sewers. My approach would be to do this in small steps, beginning by diverting some storm runoff into small wetlands or French drain retention. Separating the sewers this way is cheaper and more environmentally sound than trying to treat the mixed flow in Detroit, and the wetlands and drains would provide pleasant park spaces, but the project will take decades to complete. If done right, this would save quite a lot over sending so much liquid to Detroit, and it’s the real solution to worries about your floor drains back-flowing toxic sludge into your basement.

The incumbent, I fear, has little clue about drainage or bio-treatment. His solution is to build a $40MM tunnel cistern along Middlebelt road. This cistern only holds 3 MM gallons, less than 1/100 of the volume needed for even a moderate rain. Besides, at $13/gallon of storage, it is very costly solution compared to my preference — a French drain (costs about 25¢/gallon of storage). The incumbents cistern has closed off traffic for months between 12 and 13 mile, and is expected to continue for a year, until January, 2017. It doesn’t provide any bio-cleaning, unlike a French drain, and the cistern leaks. Currently groundwater is leaking in. This has caused the lowering of the water table and the closure of private wells. If the leak isn’t fixed , the cistern will leak septic sewage into the groundwater, potentially infecting people for miles around with typhus, cholera, and all sorts of 3rd world plagues.

There is also an explosion hazard to the incumbent’s approach. A tunnel cistern like this blew up near my undergraduate college sending manhole covers flying. This version has much bigger manhole covers: 15′ cement, not 2′ steel. If someone pours gasoline down the drain during a rainstorm and if a match went in later, the result could be deadly. The people building these projects are the same ones who fund the incumbent’s campaign, and I suspect they influenced him for this mis-chosen approach. They are the folks I fear he goes to for engineering advice. If you’d like to see a change for the better. Elect me, Elect an engineer.

Dr. Robert E. Buxbaum, March 26, 2016. Go here to volunteer or contribute.

How to help Flint and avoid lead here.

Leave a reply

As most folks know, Flint has a lead-poisoning problem that seems to have begun in April, 2014 when the city switched its water supply from Detroit-supplied, Lake Huron water to their own source, water from the Flint River. Here are some thoughts on how to help the affected population, and how to avoid a repeat in Oakland county, where I’m running for water commissioner. First observation, it is not enough to make sure that the source water does not contain lead. The people who decided on the switch had found that the Flint river water had no significant content of lead or other obvious toxins. A key problem, it seems: the river water did not contain anticorrosion phosphates, and none, it seems, were added by the Flint water folks. It also seems that insufficient levels of chlorine (hypochlorite) were added. After the switch, citizens started seeing disgusting, brown water come from their taps, and citizens with lead pipes or solder were poisoned with ppb-levels of lead. There was also an outbreak of legionaries disease that killed 12 people. It was the legionaries that alerted the CDC to the possibility of lead, since it seems the water folks were fudging the numbers there, and hiding that part of the problem.

Flint water, Sept 2015, before switching back to Lake Huron.

Flint water after 5 hours of flushing, Sept 2015, before switching back to Lake Huron.

The city began solving its problem by switching back to Detroit-supplied, Lake Huron water in October, 2015. Beginning in December, 2015, they started adding triple doses of phosphate to the wate. As a result, Flint tap-water is now back within EPA standards, but it’s still fairly unsafe, see here for more details.

There has been a fair amount of finger-pointing. At Detroit for raising the price of water so Flint had to switch, at water officials ignoring the early signs of lead and fudging their reports, at other employees for not adding phosphate or enough chlorine, and at “the system” for not providing Flint’s government with better oversight. My take is that a lot of the problem came from the ignorance of the water commission, and it’s commissioner. We elect our water commissioners to be competent overseers of complex infrastructure, but in may counties folks seem to pick them the same way they pick aldermen: for a nice smile, a great handshake, and an ability to remember names. That, anyway, seems to be the way that Oakland got its current water commissioner. When you pick your commissioner that way, it’s no surprise that he (or she) isn’t particularly up on corrosion chemistry, something that few people understand, and fewer care about until it bites them.

Flint river water contains corrosive chloride that probably helped dissolve the lead from pipes and solder. Contributing to the corrosion problem, I’m going to guess that Flint River water also contains, relatively little carbonate, but significant amounts of chelating chemicals, like EDTA, in 10s of ppb concentration. EDTA isn’t poisonous at these concentrations, but it’s common in industry and is the most commonly used antidote for lead poisoning. EDTA extracts lead and other metals from people and would tend to contribute to the process of extracting lead and iron oxide from the pipes surface into the drinking water. With EDTA in the water, a lot of phosphate or hypochlorite would be needed to avoid the lead poisoning problem and the deadly multiplication of disease.

Detroit ex-mayor Kwame Kilpatrick has claimed that both Flint water and Detroit water were known to be poisoned even a decade before the switch. I find these claims believable given the high levels of lead in kids blood even before the switch. Also, I note that there are areas of Detroit where the blood-lead levels are higher than Flint. Flint tested at the taps in a way that fudged the data during the first days of the poisoning, and I suspect many of our MI cities do this today — just to make the numbers look better. My first suggestion therefore is to test correctly, both at the pipes and at the taps; lead pipes are most-often found in the last few feet before the tap. In particular, we should test at all schools and other places where the state has direct authorization to fix the problem. A MI senate bill has been proposed to this effect, but I’m not sure where it stands in the MI house. It seems there are movements to add lots of ‘riders’ and that’s usually a bad sign.

Another thought is that citizens should be encouraged to test their private taps and helped to fix them. The state can’t come in and test or rip out your private pipes, even if they suspect lead, but the private owner has that authorization. The state could condemn a private property where they believe the water is bad, but I doubt they could evict the residents. It’s a democratic republic, as I understand; you have the right to be deadly stupid. But I’ll take my own suggestion to encourage you: If you think your water has lead, take a sample and call (517) 335-8184. Do it.

Another suggestion, perhaps the easiest and most important, is drink bottled water for now, and if you feel you’ve been poisoned, take an antidote.  As I understand things, the state is already providing bottles of imported water. The most common antidote is, as I’d mentioned, EDTA. Assuming that Flint River water had enough EDTA to significantly worsen the problem, the cheapest antidote might be Flint River water, assuming you drew it in lead-free pipes and chlorinated sufficiently to rid it of bugs. If there is EDTA it will help the poisoned. Another antidote is Succinic acid, something sold by REB Research, my company. As with EDTA it is non-toxic, even in fairly large doses, but its use would have to be doctor- approved.

Robert E. Buxbaum, January 19-31, 2016. I hope this helps. We’d have to check Flint River water for levels of EDTA, but I suspect we’d find biologically significant concentrations. If you think Oakland should have an engineer in charge of the water, elect Buxbaum for water commissioner.

Why are glaciers blue

Leave a reply

i recently returned from a cruse trip to Alaska and, as is typical for such, a highlight of the trip was a visit to Alaska’s glaciers, in our case Hubbard Glacier, Glacier bay, and Mendenhall Glacier. All were blue — bright blue, as were the small icebergs that broke off. Glacier blocks only 2 feet across were bright blue like the glaciers themselves.

Hubbard Glacier, Alaska. Note how blue the ice is

Hubbard Glacier, Alaska. My photo. Note how blue the ice is

What made this interesting/ surprising is that I’ve seen ice sculptures that are 5 foot thick or more, and they are not significantly blue. They have a very slight tinge, but are generally more colorless than glass to my ability to tell. I asked the park rangers why the glaciers were blue, but was given no satisfactory answer. The claim was that glacier ice contained small air bubbles that scattered light the same way that air did. Another park ranger claimed that water is blue by nature, so of course the glaciers were too. The “proof” to this was that the sea was blue. Neither of these seem quite true to me, though there seamed some grains of truth. Sea water, I notice, is sort of blue, but isn’t this shade of blue, certainly not in areas that I’ve lived. Instead, sea water is a rather grayish similar to mud and sea-weeds that I’d expect to find on the sea floor. What’s more, if you look through the relatively clear water of a swimming-pool water to the white-tile bottom, you see only a slight shade of blue-green, even at the 9 foot depth where the light you see has passed through 18 feet of water. This is far more water than an iceberg thickness, and the color is nowhere near as pure blue and the intensity nowhere near as strong.

Plymouth, MI Ice sculpture -- the ice is fairly clear, like swimming pool water

Plymouth, MI Ice sculpture — the ice is fairly clear, like swimming pool water

As for the bubble explanation, it doesn’t seem quite right, either. The bubble size would be non-uniform, with many quite large resulting in a mix of scattered colors — an off white– something seen with the sky of mars. Our earth sky is a purer blue, but this is not because of scattering off of ice-crystals, dust or any other small particles, but rather scattering off the air molecules themselves. The clear blue of glaciers, and of overturned icebergs, suggests (to me) a single-size scattering entity, larger than air molecules, but much smaller than the wavelength of visible light. My preferred entity would be a new compound, a clathrate structure compound, that would be formed from air and ice at high pressures.

An overturned ice-burg is remarkably blue: far bluer than an Ice sculpture. I claim clathrates are the reason.

An overturned ice-burg is remarkably blue: far bluer than an Ice sculpture. I claim clathrates are the reason.

Sea-water forms clathrate compounds with natural gas at high pressures found at great depth. My thought is that similar compounds form between ice and one or more components of air (nitrogen, oxygen, or perhaps argon). Though no compounds of this sort have been quite identified, all these gases are reasonably soluble in water so that suggestion isn’t entirely implausible. The clathrates would be spheres, bigger than air molecules and thus should have more scattering power than the original molecules. An uneven distribution would explain the observation that the blue of glaciers is not uniform, but instead has deeper and lighter blue edges and stripes. Perhaps some parts of the glacier were formed at higher pressures one could expect that these would form more clathrate compounds, and thus more blue. One sees the most intense blue in overturned icebergs — the parts that were under the most pressure.

Robert Buxbaum, October 12, 2015. By the way, some of Alaska’s glaciers are growing and others shrinking. The rangers claimed this was the bad effect of global warming: that the shrinking glaciers should be growing and the growing ones shrinking. They also worried that despite Alaska temperatures reaching 40° below reasonably regularly, it was too warm (for whom?). The lowest recorded temperature in Fairbanks was -66°F in 1961.

Thermodynamics of hydrogen generation

7 Replies

Perhaps the simplest way to make hydrogen is by electrolysis: you run some current through water with a little sulfuric acid or KOH added, and for every two electrons transferred, you get a molecule of hydrogen from one electrode and half a molecule of oxygen from the other.

2 OH- –> 2e- + 1/2 O2 +H2O

2H2O + 2e- –>  H2 + 2OH-

The ratio between amps, seconds and mols of electrons (or hydrogen) is called the Faraday constant, F = 96500; 96500 amp-seconds transfers a mol of electrons. For hydrogen production, you need 2 mols of electrons for each mol of hydrogen, n= 2, so

it = 2F where and i is the current in amps, and t is the time in seconds and n is the number electrons per molecule of desired product. For hydrogen, t = 96500*2/i; in general, t = Fn/i.

96500 is a large number, and it takes a fair amount of time to make any substantial amount of hydrogen by electrolysis. At 1 amp, it takes 96500*2 = 193000 seconds, 2 days, to generate one mol of hydrogen (that’s 2 grams H2 or 22.4 liters, enough to fill a garment bag). We can reduce the time by using a higher current, but there are limits. At 25 amps, the maximum current of you can carry with house wiring it takes 2.14 hours to generate 2 grams. (You’ll have to rectify your electricity to DC or you’ll get a nasty H2 /O2 mix called Brown’s gas, While normal H2 isn’t that dangerous, Browns gas is a mix of H2 and O2 and is quite explosive. Here’s an essay I wrote on separating Browns gas).

Electrolysis takes a fair amount of electric energy too; the minimum energy needed to make hydrogen at a given temperature and pressure is called the reversible energy, or the Gibbs free energy ∆G of the reaction. ∆G = ∆H -T∆S, that is, ∆G equals the heat of hydrogen production ∆H – minus an entropy effect, T∆S. Since energy is the product of voltage current and time, Vit = ∆G, where ∆G is the Gibbs free energy measured in Joules and V,i, and t are measured Volts, Amps, and seconds respectively.

Since it = nF, we can rewrite the relationship as: V =∆G/nF for a process that has no energy losses, a reversible process. This is the form found in most thermodynamics textbooks; the value of V calculated this way is the minimum voltage to generate hydrogen, and the maximum voltage you could get in a fuel cell putting water back together.

To calculate this voltage, and the power requirements to make hydrogen, we use the Gibbs free energy for water formation found in Wikipedia, copied below (in my day, we used the CRC Handbook of Chemistry and Physics or a table in out P-chem book). You’ll notice that there are two different values for ∆G depending on whether the water is a gas or a liquid, and you’ll notice a small zero at the upper right (∆G°). This shows that the values are for an imaginary standard state: 20°C and 1 atm pressure. You can’t get 1 atm steam at 20°C, it’s an extrapolation; behavior at typical temperatures, 40°C and above is similar but not identical. I’ll leave it to a reader to send this voltage as a comment.

Liquid H2O formation ∆G° = -237.14
Gaseous H2O formation ∆G° = -228.61

The reversible voltage for creating liquid water in a reversible fuel cell is found to be -237,140/(2 x 96,500) = -1.23V. We find that 1.23 Volts is about the minimum voltage you need to do electrolysis at 0°C because you need liquid water to carry the current; -1.18 V is about the maximum voltage you can get in a fuel cell because they operate at higher temperature with oxygen pressures significantly below 1 atm. (typically). The minus sign is kept for accounting; it differentiates the power out case (fuel cells) from power in (electrolysis). It is typical to find that fuel cells operate at lower voltages, between about .5V and 1.0V depending on the fuel cell and the power load.

Most electrolysis is done at voltages above about 1.48 V. Just as fuel cells always give off heat (they are exothermic), electrolysis will absorb heat if run reversibly. That is, electrolysis can act as a refrigerator if run reversibly. but electrolysis is not a very good refrigerator (the refrigerator ability is tied up in the entropy term mentioned above). To do electrolysis at reasonably fast rates, people give up on refrigeration (sucking heat from the environment) and provide all the entropy needed for electrolysis in the electricity they supply. This is to say, they operate at V’ were nFV’ ≥ ∆H, the enthalpy of water formation. Since ∆H is greater than ∆G, V’ the voltage for electrolysis is higher than V. Based on the enthalpy of liquid water formation,  −285.8 kJ/mol we find V’ = 1.48 V at zero degrees. The figure below shows that, for any reasonably fast rate of hydrogen production, operation must be at 1.48V or above.

Electrolyzer performance; C-Pt catalyst on a thin, nafion membrane

Electrolyzer performance; C-Pt catalyst on a thin, nafion membrane

If you figure out the energy that this voltage and amperage represents (shown below) you’re likely to come to a conclusion I came to several years ago: that it’s far better to generate large amounts of hydrogen chemically, ideally from membrane reactors like my company makes.

The electric power to make each 2 grams of hydrogen at 1.5 volts is 1.5 V x 193000 Amp-s = 289,500 J = .080 kWh’s, or 0.9¢ at current rates, but filling a car takes 20 kg, or 10,000 times as much. That’s 800 kW-hr, or $90 at current rates. The electricity is twice as expensive as current gasoline and the infrastructure cost is staggering too: a station that fuels ten cars per hour would require 8 MW, far more power than any normal distributor could provide.

By contrast, methanol costs about 2/3 as much as gasoline, and it’s easy to deliver many giga-joules of methanol energy to a gas station by truck. Our company’s membrane reactor hydrogen generators would convert methanol-water to hydrogen efficiently by the reaction CH3OH + H2O –> 3H2 + CO2. This is not to say that electrolysis isn’t worthwhile for lower demand applications: see, e.g.: gas chromatography, and electric generator cooling. Here’s how membrane reactors work.

R. E. Buxbaum July 1, 2013; Those who want to show off, should post the temperature and pressure corrections to my calculations for the reversible voltage of typical fuel cells and electrolysis.

My steam-operated, high pressure pump

3 Replies

Here’s a miniature version of a duplex pump that we made 2-3 years ago at REB Research as a way to pump fuel into hydrogen generators for use with fuel cells. The design is from the 1800s. It was used on tank locomotives and steamboats to pump water into the boiler using only the pressure in the boiler itself. This seems like magic, but isn’t. There is no rotation, but linear motion in a steam piston of larger diameter pushes a liquid pump piston with a smaller diameter. Each piston travels the same distance, but there is more volume in the steam cylinder. The work from the steam piston is greater: W = ∫PdV; energy is conserved, and the liquid is pumped to higher pressure than the driving steam (neat!).

The following is a still photo. Click on the YouTube link to see the steam pump in action. It has over 4000 views!

Mini duplex pump. Provides high pressure water from steam power. Amini version of a classic of the 1800s Coffee cup and pen shown for scale.

Mini duplex pump. Provides high pressure water from steam power. A mini version of a classic of the 1800s Coffee cup and pen shown for scale.

You can get the bronze casting and the plans for this pump from Stanley co (England). Any talented machinist should be able to do the rest. I hired an Amish craftsman in Ohio. Maurice Perlman did the final fit work in our shop.

Our standard line of hydrogen generators still use electricity to pump the methanol-water. Even our latest generators are meant for nom-mobile applications where electricity is awfully convenient and cheap. This pump was intended for a future customer who would need to generate hydrogen to make electricity for remote and mobile applications. Even our non-mobile hydrogen is a better way to power cars than batteries, but making it mobile has advantages. Another advance would be to heat the reactors by burning the waste gas (I’ve been working on that too, and have filed a patent). Sometimes you have to build things ahead of finding a customer — and this pump was awfully cool.

Metals and nonmetals

6 Replies

Hydrogen is both a metal an a non-metal. It says so on the specially produced coffee cups produced by my company (and sold by my company) but not on any other periodic table i’ve seen. That’s a shame for at least two reason. First, on a physiochemical level, while hydrogen is a metal in the sense that it combines with non-metals like chlorine and oxygen to form HCl and H2O, it’s not a metal in how it looks (not very shiny, malleable, etc.). Hydrogen acts like a chemical non-metal in the sense that it reacts with most metals to form metal hydrides like NaH CaH2 and YH3 (my company sells metal hydride getters, and metal membranes that use this property), and it also looks like a non-metal; it’s a gas like non-metallic chlorine, fluorine, and oxygen.

REB Research, Periodic table coffee cup

REB Research, Periodic table coffee cup

Most middle schoolers and high schoolers learn to differentiate metals and nonmetals by where they sit on the periodic tables they are given, and by general appearance and feel, that is by entirely non-scientific methods. Most of the elements on the left side of their periodic tables are shiny and conduct electricity reasonably well, so students come to believe that these are fundamental properties of metals without noting that boron and iodine (on the right side) are both shiny and conduct electricity, while hydrogen (presumably the first metal) does not. Students note that many metals are ductile without being told that calcium and chromium are brittle, while boron and tin (non-metals) are ductile. And what’s with the jagged dividing line: some borderline cases, like aluminum, look awfully metallic by normal standards.

The actual distinction, and the basis for the line, has nothing to do with the descriptions taught in middle school, but everything to do with water. When an element is oxidized to its most common oxide and dissolved in water the solution will be either acidic or basic. This is the basis of the key distinction: we call something a metal if the metal oxide solution is basic. We call something a non-metal if the oxide solution is an acid. To make sulfuric acid or nitric acid: you dissolve the oxides of sulfur or nitrogen respectively, in water. That’s why nitrogen and sulfur are nonmetals. Similarly, since you make boric acid by dissolving boron oxide in water boron is a non-metal. Calcium is a metal because calcium oxide is lime, a strong base. Aluminum and antimony are near borderline cases, because their oxides are nearly neutral.

And now we return to hydrogen and my cup. hydrogen is the only element listed as both a metal and a non-metal because hydrogen oxide is water. It is entirely neutral. When water dissolves in water the pH is 7; by definition, hydrogen is the only real borderline case. It is not generally shown that way, but it is shown as a metal and a non metal is on a cup produced by my company.