Researchers Propose Silica Microspheres as Stratospheric Solar Geoengineering Method

In a lab south of Tel Aviv, engineers at a startup called Stardust Solutions are manufacturing particles smaller than a red blood cell — amorphous silica microspheres mixed with calcium carbonate — designed to be lofted into the stratosphere by the millions of tons [1]. Their stated goal: reflect enough sunlight back into space to measurably cool the Earth. The company has raised $75 million, plans outdoor experiments beginning in 2026, and has published a blueprint describing the industrial infrastructure needed to deploy 10 million metric tons of engineered particles per year for what it calls "1% solar-flux modification" [2][3].

The proposal sits at the center of a collision between climate desperation and global governance failure. As temperatures continue to rise past internationally agreed targets, stratospheric aerosol injection (SAI) — the broader category of techniques that includes silica microsphere deployment — has moved from the margins of climate science to a subject of active investment, fierce opposition, and unresolved legal questions.

The Science: From Pinatubo to Engineered Particles

The basic concept behind SAI borrows from volcanic eruptions. When Mount Pinatubo erupted in the Philippines in June 1991, it injected roughly 17 million tons of sulfur dioxide into the stratosphere [4]. That SO₂ reacted with water vapor to form sulfuric acid aerosol droplets, which reflected incoming sunlight. Global temperatures dropped by approximately 0.5°C over the following 18 months — a measurable, temporary cooling that affected the entire planet [4][5].

The standard SAI proposal replicates this effect artificially, using aircraft to inject sulfur dioxide or sulfuric acid into the stratosphere on a continuous basis. Models estimate that roughly 5 million tons of sulfate aerosol per year could offset about 1°C of warming [6]. But sulfate aerosols carry known downsides: they produce sulfuric acid that damages the ozone layer, and modeling has shown that an injection large enough to offset a doubling of atmospheric CO₂ would delay recovery of the Antarctic ozone hole by 30 to 70 years [7].

This is where silica enters the picture. A 2016 paper in the Proceedings of the National Academy of Sciences by David Keith, Frank Keutsch, and colleagues at Harvard proposed that solid aerosol particles — specifically calcite (calcium carbonate) and potentially other minerals including silica — could achieve stratospheric cooling without the ozone-destroying chemistry of sulfate aerosols [8]. Calcite particles would neutralize emissions-borne acids in the stratosphere rather than generating new ones, theoretically allowing geoengineering while moving ozone concentrations back toward pre-industrial baselines [8][9].

Stardust Solutions builds on this line of research. Its particles are made from amorphous silica — a material already used as a food additive — and calcium carbonate, which occurs naturally in limestone and eggshells [3]. The company argues that these materials are inert enough to avoid the acid-chemistry problems of sulfate aerosols.

However, no peer-reviewed literature has yet confirmed the cooling efficiency of silica microspheres specifically — measured in degrees Celsius of cooling per teragram (one million metric tons) deployed — through outdoor experiments. The figures that exist are modeled, not empirically validated in the open atmosphere [10]. This is a critical gap: the difference between laboratory properties and stratospheric behavior under real-world conditions (particle clumping, sedimentation rates, atmospheric chemistry interactions) remains untested.

The Cost Equation

SAI is often described as remarkably cheap relative to the problem it addresses. A widely cited 2018 study by Wake Smith and Gernot Wagner estimated the annual cost of a sulfate-based SAI program at roughly $2 billion to $8 billion per year for delivering 5 million tons of aerosol to altitudes of 20–30 kilometers, or approximately $18 billion per year per degree Celsius of warming offset [6][11].

Source: Smith & Wagner 2018; Rising et al. 2022

Data as of Oct 1, 2023CSV

Compare that to the projected economic cost of climate change itself. Research published in One Earth in 2023 estimated annual climate damages at $19 trillion to $59 trillion by 2050 under insufficient mitigation scenarios [12]. The cost-benefit ratio appears extreme: SAI deployment costs are roughly three orders of magnitude smaller than the damages it might prevent.

But this comparison is misleading if taken at face value. The $18 billion estimate covers only the direct costs of manufacturing aerosols and flying aircraft. It does not include the cost of the global monitoring infrastructure needed to track regional climate effects, the agricultural adaptation programs required for areas where rainfall patterns shift, the diplomatic and governance apparatus needed to manage international disputes, or the indefinite commitment to continued injection (since stopping triggers its own catastrophe, discussed below) [13].

Stardust Solutions has not published a full lifecycle cost estimate for silica microsphere deployment at climate-relevant scale.

Who Gets Hurt: Regional Disparities and Monsoon Risk

SAI does not cool the Earth evenly. Models consistently show that injecting aerosols at tropical or subtropical latitudes reduces incoming solar radiation globally but alters precipitation patterns regionally — with the most severe effects concentrated in the tropics, where monsoon systems drive agricultural production for billions of people [14][15].

A 2025 study published in the International Journal of Climatology found that SAI scenarios produced overall precipitation reductions across South Asia relative to projected climate change conditions [15]. Other modeling work has projected 5 to 7 percent less annual rainfall across parts of the tropics under large-scale aerosol injection [14]. If injection is applied unevenly between hemispheres — cooling one more than the other — the impact on the Indian Summer Monsoon could be severe enough to effectively suppress it [14].

The populations most dependent on monsoon rainfall for food production — in South Asia, West Africa, and Southeast Asia — are overwhelmingly in the Global South. These are also the nations that have contributed least to cumulative greenhouse gas emissions. Multiple studies have flagged this distributional injustice: the communities most likely to suffer side effects of SAI are the same ones bearing the worst consequences of climate change itself [14][16].

This concern is not theoretical. When Switzerland proposed establishing an expert advisory panel on solar radiation modification (SRM) at the sixth UN Environment Assembly (UNEA-6) in February 2024, a coalition of African nations led by Djibouti opposed the resolution. The African group, eventually joined by Vanuatu, Fiji, Mexico, Pakistan, and Colombia, called instead for a governance mechanism mandating the "non-use" of SRM technologies [17][18]. Switzerland withdrew the resolution — the second such failure after a similar attempt in 2019 [17].

The Governance Vacuum

No international treaty specifically governs stratospheric aerosol injection. No institution has clear legal authority to permit, regulate, or ban it. No enforcement mechanism exists to prevent a single nation or private actor from deploying it unilaterally [19][20].

Existing international law offers partial coverage. The principle of preventing significant transboundary harm — a cornerstone of customary international law — applies in theory, since SAI deployed by one country would alter climate conditions in others [19]. The precautionary principle is frequently invoked. In 2024, the International Tribunal for the Law of the Sea issued an advisory opinion emphasizing that marine geoengineering could violate the UN Convention on the Law of the Sea if it introduces pollutants into the marine environment [19]. But stratospheric injection is not marine geoengineering, and no comparable ruling covers it.

The Center for International Environmental Law (CIEL) has called for strengthening and enforcing what restrictive frameworks exist, arguing that the UNEA-6 resolution failure left a vacuum that private actors are already filling [20]. Stardust Solutions, for instance, announced plans for outdoor experiments from April 2026 without seeking approval from any international body — because none has jurisdiction to grant or deny it [2][3].

The SCoPEx Precedent: Why Past Research Was Blocked

The most prominent prior attempt at outdoor SAI research — Harvard's Stratospheric Controlled Perturbation Experiment (SCoPEx) — was canceled in March 2024 after a decade of planning [21][22]. The experiment, led by professors Frank Keutsch and David Keith, would have released a small quantity of calcium carbonate particles from a balloon in northern Sweden to measure how they dispersed.

The project was abandoned not because of a scientific failure but because of organized opposition. In early 2024, a coalition of Swedish environmental organizations and the Indigenous Saami Council demanded cancellation, arguing that any outdoor geoengineering experiment — regardless of scale — would legitimize a technology that poses unacceptable risks [22][23]. The Harvard advisory committee overseeing SCoPEx had taken the "moral hazard" concern seriously: the possibility that research into a technological fix would reduce political pressure to cut emissions at their source [21].

Proponents of silica microspheres — including Stardust Solutions — argue their approach addresses the specific safety objection that helped kill SCoPEx. Because silica and calcite are chemically inert relative to sulfate aerosols, they claim the ozone-depletion risk is eliminated and the particle behavior is more predictable [3][8]. But the SCoPEx opposition was never solely about ozone chemistry. The Saami Council and allied groups objected to the entire premise of stratospheric intervention, not to the specific material being tested [23].

The scientific evidence that silica microspheres are "meaningfully safer" remains theoretical. The 2016 Keith et al. PNAS paper demonstrated through atmospheric chemistry modeling — not outdoor observation — that calcite particles could neutralize stratospheric acids [8]. No equivalent peer-reviewed study has validated this for amorphous silica microspheres specifically. The gap between "works in a model" and "works in the stratosphere" is precisely what outdoor experiments were meant to close — and those experiments keep getting canceled.

Termination Shock: The Trap of Starting

Perhaps the most sobering aspect of SAI is what happens when you stop. If aerosol injection were sustained for years or decades — masking several degrees of warming — and then halted abruptly, the accumulated greenhouse warming that had been temporarily suppressed would return rapidly. This is called termination shock [24].

Source: Emulating SAI Inconsistencies, IOP Science 2024

Data as of Jun 1, 2024CSV

A 2024 study published in Environmental Research: Climate modeled the warming rates under various termination scenarios. Under abrupt termination, temperatures rebounded at 0.88°C per decade — nearly three times the baseline warming rate of 0.31°C per decade without SAI. A 10-year phase-out scenario produced even faster peak warming of 1.0°C per decade [25]. For comparison, the rate of warming that ecosystems, agriculture, and human infrastructure have struggled to adapt to over the past century has been roughly 0.2°C per decade.

A termination shock at these rates would compress decades of expected warming into a few years. The populations most vulnerable would be those in tropical regions with limited adaptive capacity — small island states, subsistence farming communities, and coastal populations in the Global South [13][14].

This creates what critics call a "commitment trap": once started, SAI is extremely dangerous to stop, effectively locking humanity into indefinite deployment regardless of side effects, costs, or geopolitical changes [24].

Follow the Money

Stardust Solutions was founded in 2023 by Yanai Yadov, a former deputy chief scientist at the Israeli Atomic Energy Commission, along with Amiad Spektor and Eli Waxman [1][3]. The company is registered in the United States with an Israeli subsidiary and operates a lab in Ness Ziona, south of Tel Aviv [3].

The $75 million the company has raised came from multiple sources. A $60 million round drew from Silicon Valley venture capitalists and Piero Ferrari — the largest shareholder in Ferrari and the Italian football club Juventus [2]. An earlier $15 million round included investment from AWZ, an Israeli-Canadian venture capital firm that has faced criticism for its partnership with the Israeli Ministry of Defense's Directorate of Defence Research and Development [2][26].

The funding landscape on the opposition side is substantially different. The Solar Geoengineering Non-Use Agreement initiative, launched in January 2022 by more than 60 senior climate scientists and governance scholars, operates as an academic coalition without venture capital backing [27]. The initiative now counts over 500 scholars from 67 countries and more than 1,900 civil society organizations among its supporters [27].

This asymmetry matters. The proponents of silica microsphere deployment have access to tens of millions in private capital and a clear financial incentive to develop deployable technology. The opponents are largely academics and civil society groups funded through research grants and philanthropic foundations. The question of who profits from deployment — and who bears the risk — runs through the entire debate.

The Steelman Case Against

The strongest argument against solar geoengineering does not come from Luddites or climate deniers. It comes from climate scientists who accept the severity of warming but argue that SAI would make things worse in the long run. Their case rests on three pillars [27][28]:

Moral hazard is real, not hypothetical. If governments and corporations believe a technological backstop exists, the political will to cut emissions — already insufficient — will erode further. The Non-Use Agreement proponents argue that the perception of SAI availability will reduce policy commitment to emission reductions [27]. Some experimental evidence supports this concern: studies have shown that presenting people with information about geoengineering reduces their stated willingness to support carbon taxes and emission reduction policies [28].

Permanent technological dependency. SAI does not address the root cause of warming — atmospheric greenhouse gas concentrations. It treats the symptom (excess heat) while the underlying condition (excess CO₂) worsens. This means deployment cannot be temporary. Once started, it must continue indefinitely — and the required dose increases over time as CO₂ accumulates. Any interruption triggers termination shock [24][25].

Risk transfer from emitters to non-emitters. The countries that have emitted the most greenhouse gases are wealthy, temperate-latitude nations. The countries most likely to experience disrupted monsoons, altered precipitation, and reduced agricultural yields from SAI are tropical and subtropical developing nations [14][16]. SAI thus shifts climate risk from those who caused it to those who did not — without their consent and without governance mechanisms to ensure accountability [19][27].

Whether the moral hazard effect is large enough to outweigh the potential cooling benefits remains contested. A 2022 study in Climate Policy argued that the value of information from SAI research is substantial and that the "bias" of assuming moral hazard without testing it empirically carries its own costs [29]. But the burden of proof sits uneasily: proponents must demonstrate that a technology with irreversible global consequences is safe enough to test outdoors, while critics argue that the governance infrastructure to manage such testing does not yet exist.

The Research Accelerates Anyway

Source: OpenAlex

Data as of Jan 1, 2026CSV

Academic publication on solar geoengineering has surged. According to OpenAlex data, 902 papers on the topic were published in 2025, up from 114 in 2011 — an eightfold increase over 14 years [30]. The field is no longer a fringe pursuit. NOAA scientists published proposed guidelines for SAI research evaluation in August 2024 [10]. The US National Academies of Sciences recommended in a 2021 report that the US invest in a solar geoengineering research program — while stressing that research is not endorsement of deployment [31].

Meanwhile, Stardust Solutions is not waiting for international consensus. The company has been conducting indoor trials since 2022 and outdoor hardware testing since 2024 [1]. CIEL, the international environmental law organization, has called the planned April 2026 outdoor experiments "reckless" and warned that they set a precedent for unilateral action without any governance framework in place [26].

The fundamental tension has not changed since Mount Pinatubo cooled the planet for a year and a half in 1991: the atmosphere does not recognize national borders, and no institution exists that can speak for the entire planet. Whether silica microspheres are safer than sulfate aerosols is a scientific question that cannot be answered without experiments — but whether those experiments should proceed is a political and ethical question that no one has the authority to resolve.