ChemRxiv vs. arXiv acceptance

The main problem is that your impressions seem to be incorrect. (At least from my maths perspective) Putting a preprint on arxiv is just that, namely putting up a preprint. Almost everything put on arxiv will be sent of to a peer reviewed venue as well, usually at the same time (there are exceptions, e.g. when people put up non-articles such as lecture notes for posterity). It is useful to make a result widely available and in cases even to establish priority, but it is not "proper" publishing and generally is not accepted as such. It is just a way to put the result out there right now instead of waiting until it is published (which in cases can take years).

In a way this approach is an attempt to please both sides. The bean-counters and bureaucrats who care about numerical "publication record" generally ignore arxiv and look at peer reviewed stuff only. But the actual peers reading your papers care more about the content than about the venue it is published in. (Though they still will cite the published version if available)

Also keep in mind that people can only cite what they actually know about. And in some disciplines, the use of arxiv is so prevalent that it also functions as a unified source. Since every preprint is on there anyway, many people only check the new releases on arxiv, so not releasing a preprint is not an option anymore.

Clarifying note: this addresses the main thrust of question regarding preprints: even in preprint-happy disciplines, preprints are not a replacement for peer-review.

It's really just cultural, and in twenty years this will likely look like a silly question to a newer generation of researchers.

The preprint movement happened to start in physics, which may have been predisposed to it due to the use of TeX and the long lead-times of much work in both theoretical and experimental physics. The arXiv merely formalized this practice, then helped enable it to spread to other closely culturally related fields---particularly the more theoretical portions of those fields.

The next big cultural leap was bioRxiv, moving into the field of biology. The barriers to starting there were, once again, facilitated by cross-disciplinary linkages, this time with the more mathematical bioinformatics sorts of researchers, who had gained experience and comfort with preprint publishing on arXiv. For the past few years, bioRxiv has been growing explosively and as the community is embracing it, so are the journals, creating a feedback loop of growth.

With these two examples now so successful and the pressure for open science continuing to build, there is now a mass radiation of attempts at the same model for other fields, not just ChemRxiv, but also PsyArXiv, SocArXiv, EarthArXiv, engrXiv, etc. Between this ongoing explosion and other preprint models, it seems likely that preprint acceptance and availability will continue to grow rapidly throughout most of the scientific community.

In short: the difference in attitudes about preprints between physics and chemistry is a transient state that exists right now because chemists currently happen to be in the middle of a cultural transition that physics went through a couple of decades ago.

In short, preprints are newer in chemistry due to historical roadblocks.

Many chemists (including myself) have put preprints on arXiv for years - although mostly in the theoretical or physical chemistry subdisciplines.

ChemRxiv is a new development, starting only in August 2017, mostly because chemistry journals, particularly including American Chemical Society journals would consider pre-prints as "prior publication" during submission and would therefore reject such manuscripts outright.

There were several previous attempts to have pre-prints in chemistry, but when ChemRxiv was launched with the backing of the ACS, other leading world-wide chemical societies, and "buy-in" from top journals, it started to gain traction. In 2018, J. Amer. Chem. Soc. (JACS) also started accepting preprints - coming from one of the few remaining holdouts - was another strong signal of acceptance.

I find some colleagues still reluctant to submit, for fear of being "scooped" between a preprint appearing and the final manuscript acceptance.

However, I will almost always submit a preprint as the manuscript is submitted:

• It indicates priority - sometimes the review process is long and the preprint will be available for others to read and cite.
• I can cite the preprint in grant applications or project reports, since a DOI is assigned when published. Several grant reviewers have appreciated this.
• It offers a centralized "open access" version for manuscripts.
• I can discuss the work online and gain interest in the paper. I often receive comments outside the review process. (Indeed, I have found the preprint gets more useful comments than most reviews.)