By David Nicholls

The electricity issue in South Africa is the basis of a very intense debate. There are those who believe the entire solution is the creation of renewable based “mini-grids” with no national grid and those who believe the only solution is to return to the single, monopolistic, system powered by mega-coal stations, and every option in between. All these proposed scenarios are based on a technology or combination of technologies which the proponents claim are the best answer to our national problems. Unfortunately, while many technical solutions are well based, there are a number which can only be called “Phantom Solutions”.So what is a “Phantom Solution”? It is useful to first define what is a credible option. There are four key issues for a credible solution.A solution must be aimed to solve a problem, so there must be a statement of the problem we are trying to solve, without this you just have an interesting technology. A problem statement for the South African electricity crisis could be to “meet the national need for the provision of a reliable, cheap and abundant supply of electricity at an acceptable environmental impact?” This would encompass a 24/7 supply to support a growing industry and society.Given that a problem statement has been defined the second issue is that the technology solution proposed must be able to technically meet the requirement. As an example if the solution proposed was only based on Photo Voltaic (PV) systems with no dispatchable back-up or storage it could not meet the reliable requirement (the sun does go down at night!).

The third issue is that the solution must be economically viable. While this viability may only be after an initial series has been constructed to get the relevant development or learning effect it must not be based on a belief in some yet to be discovered invention.
The fourth issue for the technology to be a credible option must be that it is “politically acceptable”. This may not be at a purely national political level but could be at a different level, such as investor acceptance. 
An issue for this is if the option does not provide the opportunity for local economic involvement. An example of this may be the KarPowership bids for the RMIPPP tender. While they appear to have succeeded at the technical and commercial level against an open, performance-based specification they do not appear to allow adequate commercial involvement of local companies and this seems to be one of the factors that have led to a significant push back against them.So if these are the factors for a credible technology solution what are some examples of Phantom Solutions proposed for the South African electricity crisis? Probably one of the most serious is the belief that South Africa, due to the massive changes of the Fourth Industrial Revolution, has a lower electricity demand than has been projected in previous years and the need for “baseload power” has diminished. While this is not a technology solution exactly it does redefine the problem South Africa faces and therefore allow one to explain away the limitations of proposed solutions. The reality is that since 2007 Eskom’s constrained electricity supply due to a failure to start building new plants in the late 1990s has led to a sharp decline in investment in new industrial plant (such as mining and smelters) on which South Africa’s growth is dependent. This can be seen in these two graphs, showing that South Africa’s GDP per capita has not kept up with our emerging market competitors as our electrical consumption per capita has declined. As soon as the electrical supply is constrained (load shedding being an extreme symptom) the consumption figures are a function of supply, not demand. If one is talking of specific technologies, however, there are two which are currently seen to be very high profile as the major part of the solution, but fail the fundamental economics criteria.
The first is the belief that battery technology will provide the storage solution to unlock the intermittency challenge of PV and wind power in a net zero carbon constrained world. The issue with batteries is that while the cost of Lithium Ion batteries have fallen dramatically over the last two decades they have now reached the point where the material costs (lithium, cobalt, nickel etc) make up some 75% of the manufacturing costs, so further significant reductions seem unlikely. The best national reference for installed, grid scale, costs is the current Eskom Battery Storage System. This is quoted as having 1449MWh storage capacity and to cost R11bn. At an exchange rate of R18:$1 this leads to a storage cost of over $400/kWh. It is accepted that battery storage system costs have to get below $100/kWh, and probably to around $50/kWh, to make the PV/battery solution the lowest cost option for large scale grid applications. (If there is a belief that battery prices are falling to levels not seen before it is worth noting that the Lead-Acid battery, invented in 1859, is still the lowest cost option for batteries in cars with internal combustion engines – Li-Ion batteries are used in electric cars because of power density.)
One of the other “solutions” for which the fundamental economics do not appear to make much sense is “green hydrogen” as an energy carrier. While the technical ability for hydrogen to act as a storage and transport system is clearly there the economics have some major challenges. There is talk of replacing natural gas (methane) in current systems with “green hydrogen”. It is important to note that on a volume basis, for similar gas pressures, methane has three times the energy content of hydrogen. If one goes for liquefaction (as in Liquified Natural Gas – LNG) for transport the one third volume to energy is still applicable and the temperature to liquify hydrogen is 253oC below zero (only 20C above absolute zero) and it consumes 30% of the energy content of the hydrogen to do. In comparison methane liquifies at 162oC below zero. Therefore the option of replacing natural gas with hydrogen raises major economic challenges and also requires an increase in the infrastructure by three times (as well significant upgrades to manage the increased leakage of hydrogen because the molecules are smaller).
An example of this challenge is the use of hydrogen to power transport vehicles vs. using batteries. The efficiency losses of using batteries consist of the transmission and storage before delivering electricity to the wheels. That equates to about 76% of the electricity generated by the PV cell (?) getting to the road wheel. In the case of the hydrogen option the steps are generation -> electrolysis -> compression/liquefaction -> transport and filling -> fuel cell -> motor/wheel. This has a typical overall efficiency of about 32%.
The promoters of these kind of Phantom Solutions point to countries where they are said to have been applied successfully. In this approach the “success” of the application can be viewed very selectively.
An example of a recent quoted example of a “successful roll out of renewables” has been that seen in Vietnam. Since 2020 Vietnam has installed close to 20GW of roof top and IPP solar. What is not mentioned is how they did it and what they plan in the next ten years (their PDP VIII to 2030). To get this amazing success there was a standard feed-in-tariff of $0.0935/kWh with a twenty year PPA, or 18% higher than their domestic tariff (and four times the PV bids in the most recent South African REIPPP bid window). 

In their next plan (issued October 2022) Vietnam plans by 2030 to install some 25,000MW of new LNG plant, 12,000MW of new coal plant and 7,000MW of hydro electric plant. The plan includes only 708MW of new PV plant and some 6000MW of wind! Like the German experience this does not really support the belief that the renewables are the effective solution for a grid of similar size to ours.

Another example is that of Denmark, which is quoted to have its major energy source being wind power. This is admirable but proves very little. Denmark is very strongly linked to the neighbouring grids, being the Nordic grid (mainly nuclear and hydro) and the German grid, with its links to Poland and France. In fact it imports over 50% of the electricity that it consumes while exporting a similar amount it generates. This is a function of the regular oversupply and undersupply from its wind fleet not aligned to its demand. This example is of little use to South Africa which is not the small country inside a large, external, grid system. (Maye it could work for Lesotho?). Of course the ultimate measure of economic success is price and the household tariff for Denmark in 2022 was virtually the highest in the world, at $0.53/kWh, or some R10/kWh at the current R/$ exchange rate.

In conclusion the electricity debate is full of possible solutions to the crisis. While many are credible there are a number which are almost inherently “Phantom Solutions”. These Phantom Solutions sound excellent at solving the problem and having almost no negatives. Unfortunately they are usually fundamentally flawed and confuse the debate, with possibly catastrophic outcomes. 

It is vital that the debate really interrogates all options proposed against the actual requirements and actual potential of the option. South Africa’s electricity debate must be based on the presumption that the country cannot afford to gamble on an optimistic view again of hope that a new model will save us!