Is Ronald Ace's "solar trap" patent plausible? This newspaper article and a few others from last year discussed a patent by independent US inventor Ronald Ace. It's about a kind of absorber for solar thermal energy systems, and it's supposed to work so well that even unconcentrated sunlight can heat it up to ~1000°C (if I'm reading it right). This is Patent WO 2014133672 A1:


*

*Searchable text without figures

*PDF with figures
Is it plausible?
I'm adding an answer based on my own reading of the patent, but I hope others will chime in too.
 A: I didn't read every word but I'll address what I believe are three of the most important parts of the patent.
(1) One of the ideas for reducing radiative heat loss is reasonable
In solar thermal systems, the hot object can lose heat by emitting IR thermal radiation, and you want to stop that process. This is a big field but the approach most people use is to make your collector have high emissivity (blackbody) for sunlight and low emissivity (whitebody) for mid-IR. This traditional approach is a good idea but hard to do perfectly and inexpensively, which is why it's still the subject of ongoing research.
AFAICT, the patent suggests two alternative approaches. The first is to have a thick thermally-insulating layer on top of the sunlight-absorber which is the opposite of what you normally use: Blackbody for mid-IR, transparent for sunlight. It prevents IR radiative losses because the light keeps getting absorbed and reemmited. (Similar to the CO2 "greenhouse effect".) This is a reasonable approach, as long as the appropriate material can be found for the layer (see below).
The other approach is having angle-sensitive baffles, since sunlight is collimated and the thermal radiation is lambertian. This is reasonable on its face, but actually sunlight stops being collimated if you concentrate it a lot. That's why the patent suggests low-concentration system (100X or less, whereas traditional solar thermal uses much higher concentration). Here's the problem: Low-concentration systems make no sense for solar thermal, unless you really have no choice. If you keep the operating temperature the same but increase concentration, efficiency always goes up, because parasitic heat loss is a fixed function of operating temperature, so increasing the heat flow decreases the relative parasitic loss. That means: If you increase the acceptance angle of the baffle, while simultaneously increasing concentration by the same ratio, the system efficiency would improve. The patent purports to explain why low-concentration is actually good despite what everybody else says, but I don't believe (or perhaps don't understand) the argument.
But anyway, if you did have no choice but to make a low-concentration solar thermal system, an angle-baffle is a good idea.
(2) Using a thick visible-light-transparent fluid layer as a thermally-insulating blanket is probably a terrible idea.
The problem is that when a fluid has a temperature gradient, it gets convective cells that transfer heat extremely effectively. This is an obvious point. The patent says that the "counterflowing fluid" idea (see below) eliminates convection, but doesn't justify it, and I really don't believe it (due to Galilean relativity for example).
In addition to convection, there is also conductive heat transfer, especially if a fluid is hot (they claim 1000C).
I didn't read the patent carefully enough to say with certainty that a fluid can't work. Maybe there is some other way to eliminate convection that is actually plausible (super-viscous fluids? Index-matched junk in the liquid to impede flow?) Maybe the heat conductivity is lower than I think, and a 1-meter-thick layer is doable and is enough to insulate it well. I'll just say putting a fluid here is a strange and implausible strategy.
(3) The "counterflowing" fluid idea is nonsense
The idea here is that heat is flowing from the absorber, through the thermally-insulating fluid 'blanket', to the ambient environment, and we don't want this to happen. Therefore, let's continually pump ambient-temperature fluid onto the cold side of the blanket, and slowly flow it towards the hot side. The heat gets advectively transferred back towards the absorber, so none (or much less) enters the ambient environment. Problem solved, right?
This idea seems superficially very clever, but actually I think it's nonsense. Let's go back to basics. In a solar thermal system, we want something to stay very hot, and we want to minimize the rate of heat flow out of that very hot thing, because if more heat flows out than in then it won't maintain its temperature. Thermal conductivity says that heat always flows from warmer fluid to its neighboring colder fluid. That's the problem, and flowing the fluid does nothing to reduce this problem. Whether the heat is flowing ultimately into the cold environment, versus flowing into the cold fresh fluid that you continually flow in, doesn't matter.
We often describe heat flow in everyday life in terms of "coldness" flowing in the opposite direction. Indeed, backing up this intuition, we can mathematically reformulate the heat flow equation as a "coldness flow equation". Well, the advection is bringing coldness towards the solar absorber, which is the opposite of what we want! 
I'm sure that they have done simulations that suggest that it works. But I would bet that either they assumed such a low thermal conductivity and such a thick layer that it works whether it's "counterflowing" or not, or else they neglected to take into account the huge amount of heat that is leaving the system when the fluid at the hot side exits.
