--- title: We expected 100 rpm to mix fastest. It didn't. description: A gas-liquid CFD study of a 2000 L single-use bioreactor similar to a Cytiva XCellerex setup showed the fastest mixing at ~80 rpm and the slowest at 100 rpm. In this design, the impeller's main job is to control how far the gas plume spreads. image: https://www.p4b.io/hubfs/website/video/xcellerex-2000l-rpm.webp --- [![p4b](https://www.p4b.io/hubfs/raw_assets/public/p4b/images/p4b-logo-white.svg) ![p4b](https://www.p4b.io/hubfs/raw_assets/public/p4b/images/p4b-logo-navy.svg)](https://www.p4b.io/) Products [**pDS inserts**Large-scale conditions in your lab bioreactor](https://www.p4b.io/pds) [**Advanced CFD**Gradients, lifelines and cell experiences](https://www.p4b.io/cfd) [**DREAM-bio**Find the best geometry for your bioreactor](https://www.p4b.io/dream-bio) [**Ambr 250 vs. 2,000 L**Interactive CFD comparison of the mixing](https://www.p4b.io/compare) [Case studies](https://www.p4b.io/case-studies) [Insights](https://www.p4b.io/insights) [Events](https://www.p4b.io/events) [About](https://www.p4b.io/about) [Contact](https://www.p4b.io/contact) [Request a quote](https://www.p4b.io/contact#quote) [Home](https://www.p4b.io/) / [Insights](https://www.p4b.io/insights) CFD · 23 Oct 2026 # We expected 100 rpm to mix fastest. It didn't. A gas-liquid CFD study of a 2000 L single-use bioreactor similar to a Cytiva XCellerex setup showed the fastest mixing at ~80 rpm and the slowest at 100 rpm. In this design, the impeller's main job is to control how far the gas plume spreads. We ran a CFD study on a 2000 L single-use bioreactor system similar to a Cytiva XCellerex setup, with a side-mounted impeller and gassing from below. Aeration was 0.1 vvm, which is a bit on the high side for mammalian processes, and the stirrer speed ranged from 50 to 100 rpm. All cases were full gas-liquid simulations in Siemens M-Star CFD. The expectation was simple: higher rpm, faster mixing. The reality was different. Mixing was fastest at ~80 rpm and slowest at 100 rpm. Why? In this design, the system behaves much like an airlift. The impeller does more than mix: it controls how far the gas plume spreads. With too little spread, mixing is weak. With too much spread, the circulation breaks down. At 100 rpm, the bubbles were distributed too widely and the global circulation loop weakened, so mixing slowed. At ~80 rpm, the spread was just right, giving the strongest circulation and the fastest mixing. So there is a clear optimum. More rpm does not automatically improve mixing, and it can push the system away from that optimum. Stirring speed alone does not tell us how efficiently a compound moves through the vessel, which shows why two-phase simulations matter. This is the kind of non-linear behavior we explore with DREAM-bio, our automated CFD platform, mapping these effects before they show up in real processes. [← All insights](https://www.p4b.io/insights) Get in touch ## Let's find out how your process reacts to scale Tell us your system and target scale - we come back with a proposal for your insert set. [Request a quote](https://www.p4b.io/contact#quote)[Book a 30-min call](https://www.p4b.io/contact#meeting) ![Lena Achleitner](https://www.p4b.io/hs-fs/hubfs/website/team/lena-avatar.webp?width=84&height=84&name=lena-avatar.webp) **Lena Achleitner**Senior Bioprocess Engineer, Key Account Manager[lena.achleitner@p4b.io](mailto:lena.achleitner@p4b.io) ### Get new insights by email CFD results, case studies and new insert systems when we publish them. Unsubscribe any time. ![p4b](https://www.p4b.io/hubfs/raw_assets/public/p4b/images/p4b-logo-white.svg) Scale reliably. Develop faster. Digital and 3D-printed technologies for bioprocessing. [Follow us on LinkedIn](https://www.linkedin.com/company/p4b-gmbh/) #### Products - [pDS inserts](https://www.p4b.io/pds) - [Advanced CFD](https://www.p4b.io/cfd) - [DREAM-bio](https://www.p4b.io/dream-bio) - [Request a quote](https://www.p4b.io/contact#quote) #### Company - [About us](https://www.p4b.io/about) - [Case studies](https://www.p4b.io/case-studies) - [Insights](https://www.p4b.io/insights) - [Events](https://www.p4b.io/events) - [Contact](https://www.p4b.io/contact) #### Contact **Office & deliveries** p4b GmbH Leopold-Ungar-Platz 2 Spaces Square One, Stiege 2 1190 Vienna, Austria [+43 677 61062225](tel:+4367761062225) [office@p4b.io](mailto:office@p4b.io) Registered office: Gärtnersteg 4, 2120 Obersdorf, Austria Supported by ![austria wirtschaftsservice](https://www.p4b.io/hubfs/raw_assets/public/p4b/images/logos/aws.png) ![Bundesministerium Arbeit und Wirtschaft](https://www.p4b.io/hubfs/raw_assets/public/p4b/images/logos/bmaw.png) ![Bundesministerium Klimaschutz, Umwelt, Energie, Mobilität, Innovation und Technologie](https://www.p4b.io/hubfs/raw_assets/public/p4b/images/logos/bmk.png) ![FFG](https://www.p4b.io/hubfs/raw_assets/public/p4b/images/logos/ffg.png) © 2026 p4b GmbH · FN 611768i · ATU79846712 [Imprint](https://www.p4b.io/imprint)[Privacy policy](https://www.p4b.io/privacy-policy)Cookie settings ```json { "@context" : "https://schema.org", "@type" : "Organization", "address" : { "@type" : "PostalAddress", "addressCountry" : "AT", "addressLocality" : "Obersdorf", "postalCode" : "2120", "streetAddress" : "Gärtnersteg 4" }, "description" : "Scale-down technology for bioprocess development: 3D-printed pDS inserts, bioreactor CFD and DREAM-bio design optimisation.", "email" : "office@p4b.io", "logo" : "//143831650.fs1.hubspotusercontent-eu1.net/hubfs/143831650/raw_assets/public/p4b/images/p4b-logo-navy.svg", "name" : "p4b GmbH", "sameAs" : [ "https://www.linkedin.com/company/p4b-gmbh/" ], "telephone" : "+43 677 61062225", "url" : "https://www.p4b.io" } ``` ```json { "@context" : "https://schema.org", "@type" : "BlogPosting", "author" : { "@type" : "Organization", "name" : "p4b GmbH", "url" : "https://p4b.io" }, "dateModified" : "2026-10-23", "datePublished" : "2026-10-23", "description" : "A gas-liquid CFD study of a 2000 L single-use bioreactor similar to a Cytiva XCellerex setup showed the fastest mixing at ~80 rpm and the slowest at 100 rpm. In this design, the impeller's main job is to control how far the gas plume spreads.", "headline" : "We expected 100 rpm to mix fastest. It didn't.", "image" : "https://143831650.fs1.hubspotusercontent-eu1.net/hubfs/143831650/website/video/xcellerex-2000l-rpm.webp", "mainEntityOfPage" : "https://www.p4b.io/insights/faster-stirring-not-faster-mixing", "publisher" : { "@type" : "Organization", "logo" : { "@type" : "ImageObject", "url" : "//143831650.fs1.hubspotusercontent-eu1.net/hubfs/143831650/raw_assets/public/p4b/images/p4b-logo-navy.svg" }, "name" : "p4b GmbH" } } ``` ```json { "@context" : "https://schema.org", "@type" : "BlogPosting", "author" : { "@type" : "Person", "name" : "p4b team", "url" : "https://www.p4b.io/insights/author/p4b-team" }, "dateModified" : "2026-10-23T08:00:00.947Z", "datePublished" : "2026-10-23T06:08:16.000Z", "headline" : "We expected 100 rpm to mix fastest. It didn't.", "mainEntityOfPage" : { "@id" : "https://www.p4b.io/insights/faster-stirring-not-faster-mixing", "@type" : "WebPage" }, "publisher" : { "@type" : "Organization", "logo" : { "@type" : "ImageObject", "url" : "https://www.p4b.io/hubfs/website/email/p4b-logo-navy.png" } } } ```