Cleaning PV modules: what are the options?
Dust(1, 13), dirt (1), bird droppings (1), pollen (18), air pollution (15), wildfires’ ash, snow (9), and algae depositions (14) decrease PV panels’ efficiency. Studies show that dust accumulation on PV panels drops power between 20% and 50% (1). Icephobic PV panels can generate 85% more power than snow-covered ones (9).
When engineers fight for percentual points gains in efficiency of PV modules, it makes no sense to negate such returns to PV panel soiling.
Keeping PV panels clean is essential — what are the options to clean PV panels?
The first step is system design: tilt the modules at a minimum of 10º. After ensuring a minimum tilt, several solutions exist and divide into two groups: active and passive (1). Active solutions need a power source to work, while passive solutions don’t.
Active solutions are plentiful:
- Mechanical (10) (11) (16),
- Fluids (2);
- Mechanical with fluids (air, water, and others)(3)(4)(12)(17);
- Electrostatic cleaning systems (5);
- Acoustics cleaning systems (6).
Passive solutions rely on coatings applied on the panels that make stop dust, dirt, or snow accumulation. Both interact with water in opposite ways — either embracing or repelling it. Thus passive solutions subdivide into three categories:
- Super-hydrophilicity (embracing water)(7);
- Super-hydrophobic (repelling water)(8);
- Icephobic (ice repellent)(9).
The systems vary in efficiency and have potential issues with weight, scratching, resource, and energy use.
The weight of cleaning apparatus on PV modules can lead to damage. Particularly, stress loads on the PV modules can lead to micro-cracks (19). Repetitive stress loads from regular weights bearing pressure on the panels can widen the microcracks, leading to further performance loss. Maybe acoustic cleaning systems may cause micro-cracks. I have seen no mention of this possibility, but acoustic systems clean through vibrating the PV modules.
Scratching the PV panel’s glass from the scrubbing from brushes or similar apparatus can also lead to performance loss (1).
Resource and energy use also need accounting. Water use is of explicit concern: some estimates the use of 37 billion liters (~10 billion US gallons) to clean the current world PV installed capacity (13). It is enough to supply 1 million people.
Water consumption to clean PV modules also entails a significant energy footprint in pumping and transport. Notably, considering the location of several major PV farms in dry regions, optimizing annual sun hours requires importing water over long distances.
In addition, as climate change provokes increasing aridification, human and agricultural use becomes more critical. Also, available water sources will become more distant, particularly in inland locations, and desalination makes up a larger share of sweet water. And to clean solar modules, water purity is paramount (5) to avoid dirt deposition dissolved in the water.
Energy use alone is also a point to consider. Active cleaning systems’ gains in loss avoidance in PV generation must measure against their energy consumption. The more energy the cleaning system consumes, the lower the net returns from cleaning the PV modules (1).
For example, systems that use water (or other liquids) increase the system’s overall energy consumption, reducing the net returns by adding pumping. Also, more complex systems will result in more energy consumption and maintenance (1).
Hence, ideally, passive systems development seems more promising: no moving parts, no danger of scratching, or increased load bearing. However, the performance of most such systems relies on rain availability: They are more suitable for places where rain is reasonably expectable.
In dryer places, inducing dirt sliding and photodegradation of dirt (7) are potential pathways and electrostatic cleaning systems (1). However, most electrostatic cleaning works best with higher air humidities. Even for the best electrostatic systems, performance degrades below 30% air humidity (13). Places like the Arabian peninsula, with air humidities commonly below 30%, may opt for active systems (11).