Particle size less than 2.5 micrometers in diameter is more harmful to human beings cause particles can affect both your lungs and your heart.

Smaller fractions of the PM spectrum reaches further into the airways and lungs, causing more damage. However, other sections of the pollution spectrum might be equally or potentially more dangerous.

Article Particulate matter beyond mass: Recent health evidence on th...

Check out this study https://www.sciencedirect.com/science/article/abs/pii/S2214158820300350

dissolved mercury and arsenic...which enter the brain..

Unfortunately no governments in the world care to address.

your research will never see the light of day. i spent 18 years trying to solve this problem and not one government on the planet thought it was an important issue!

Don Wilson

It depends on dose and bioavailibilty !

Google for "WHO" plus "particulate matter" and see what it says about your country

Apoorva... Regarding the size of particles, the smaller the particle size the greater will be the damage the cause to the living bodies. For the chemical composition, the presence of silica and asbestos particles exerts destruction of the respiratory vessels and the whole system damage. An appreciable hazard of the smoke come from the micro particles suspended in it. Good luck

Particulate matter 2.5 (PM2.5) are more dangerous because they can get into the deep parts of your lungs — or even into your blood.

which is based on particle size.

Carl Alexander Frisk, Enyoh Christian Ebere Thank you for the articles.

Nagalakshmi Radhakrishnan Mohamed Ben-Haddad Sraboni Sarker Ashish Gupta Harry ten Brink Thank you for your suggestions.

PM 2.5 is the most harmful because they can get into the deep parts of your lungs.

Niraj Goswami Thanks

Article IMPACT OF COVID-19 INDUCED LOCKDOWN ON THE AIR QUALITY IN TH...

Dear researcher, I would suggest you to study PM2.5 and PM10, they have greater impact on human health. Please read the paprers related to them (eg: Baral and Thapa, 2020; Wang et al. 2016; Yu et al. 2017; Cheu et al. 2016, 2017; Huwang et al. 2016 and references therewith). Thanks. :)