Miniaturization for Chemical Analysis and Synthesis Andreas Manz Imperial College London 

No content

why minitiaturize volume of 1µL 1nL 1pL (1mm)3 (100µm)3 (10µm)3 600,000,000 600,000 600 25 / cm2 2500 / cm2 250 ,000/ cm2 17 min 10s 100ms 1.5 /min / cm2 250 /s / cm2 2,500,000 /s / cm2 # molecules (1nM solution) # volumes In array diffusion time # reactions (diffusion controlled) is a cube of 

10 fold miniaturization 100 x faster reactions / bioassays 100 x faster separation 1000 x smaller volume 10 x lower reagent consumption 

human perception < 1 cm is small > 10 m is big < 100 ms is immediate > 1 min is slow factor 100: 100 ms to 1 ms: not impressive 17 h to 10 min: makes a difference 10 min to 6 s: very impressive 

established semiconductor fabrication techniques light-source mask (from DWL) photo-resist on substrate developing, etching 3-dimensional structure bonding sealed microfluidic device microfabrication technique 

on-line phosphate analyser E.Verpoorte, A.Manz, H.M.Widmer, B.van der Schoot, N.F. de Rooij,Transducers ‘93, Tokyo (ISBN 4-9900247-2-9), pp 939-942 (1993). m-TAS miniaturized total analysis system 

drug discovery • first step to find new active molecules • composed of – synthesis of new compound – isolation, characterisation – bioassay • a significant effort in pharmaceutical industry, involving new technologies 

COMBINATORIAL CHEMISTRY QUALITY CONTROL BIOASSAYS 100 educts A 100 educts B 10,000 products AB 10,000 assays 10,000 assays ??? 10,000 products AB 

CHEMICAL MICROPROCESSOR SYNTAS m educt A educt B is this a hit? yes/no specific reaction specific bioassay 

chemical reaction batch Time continuous flow Length 

Continuous flow Batch process Volume given by time Volume is volume of the vessel Easy up-scale / down scale Up-scale / down-scale difficult [scaling laws!] Fluid handling inherently there Robotics, valve switching or manual Interfacing of components easy: volume flow rate is the only parameter Interfacing of components may need wait times / dummy loops Different reaction times needed can be achieved by difference in length / cross-section The slowest step in the sequence defines the rhythm of switching, if multiple samples have to be processed 

Continuous flow Batch process Ratio of timing cannot be changed without changing the channel hardware Good flexibility in change of timing Not all analysis / synthesis methods are available All analysis / synthesis methods available Band broadening is critical, if different samples should travel through the system in sequence Different samples = separate batches Problem: to create a sequence of samples is a batch process!!! 

A B C A B , A , B A B A B C , A B , C A B C A B C fluorescence detection bioassay synthesis step 1 synthesis step 2 separation separation continuous flow 

A A A A A A A B B B B B A B , A , B A B A +B A +B solvent solvent solvent solvent solvent R E A C TO R S S E P A R A T O R S S T O R A G E 

pressure induced flow local minimum for bandbroadening defines optimum flow rate How about a sequence of injected samples? 

electroosmotic flow minimum for bandbroadening at maximum speed 

pressure electroosmosis 

pressure induced flow p e a k l e n g t h / w i d t h r a t i o i s 2 0 0 : 1 t o 2 , 0 0 0 : 1 peak length / width ratio can be very small electroosm otic flow 

No content

double stranded DNA separation 

reaction with intercalating dye 

x x x x x x x x x SYBR green x x x x x x x x x x x x x double stranded DNA SYBR green x x x x x x x x x x x x x x x x x x x x x x x x SYBR green complex [fluorescing] x x x x 1) 2) 3) 

chemical reaction • In the most simple case, a molecule A meets a molecule B and reacts to give AB • many reactions are diffusion controlled • reaction time of hours in conventional lab • reaction time of 30 min in micro well plate 

R.Srinivasan, S.L.Firebaugh, I.M.Hsing, J.Ryley, M.P.Harold,K.F.Jensen, M.A.Schmidt [Massachusetts Institute of Technology and DuPont, Wilmington], chemical performance and high temperature characterization of micromachined chemical reactors, Transducers 97, Vol.1,p163-166 (1997) Figure 1: Schematic diagram of the microreactor [9] with Pt heaters and temperature sensors represented as electrical resistors. 

Figure 4. Separation of several amino acids using post-column derivatization for detection. D.J.Harrison, K.Fluri,N.Chiem, T.Tang,Z.Fan University of Alberta, Edmonton,Canada Transducers’95, Proc., vol.1, pp752-755 (1995) 

Y -shaped junction: 1:1 fluorescein-to-rhodamine B flowrate ratio (0.5 : 0.5 mL/min) 

No content

Mixing – Diffusion times D d t 2 2  Before laminar mixing D n d t 2 2 2  After laminar mixing n = number of branches, d = tubing diameter, D= diffusion coefficient 

Distributive Micromixing Device: Schematic F. G. Bessoth, A. J. de Mello and A. Manz, Anal. Commun., 1999, 36, 213-215 

16 channels 256x faster ! 

Distributive Micromixing Chip 

F. G. Bessoth, A. J. de Mello and A. Manz, Anal. Commun., 1999, 36, 213-215 Chip manifold volume 600 nL Observation channel 530 nL Distributive Micromixing Device 

fused silica capillary glue glass Si glass 

6 ms 14 ms 38 ms 94 ms 54 ms 78 ms 0 ms 

Fluorescein and Rhodamine B; Flow rate = 50 mL min-1; Time from point of confluence to beginning of long channel = ca. 9 ms laminar flow visualisation 

fast fluorescence quenching 0 ms 6 ms 

Mixing * + further downstream reaction incomplete reaction incomplete reaction complete reaction complete reaction complete 

horseradish peroxidase assay 0 1 2 3 4 5 6 7 8 0 0.02 0.04 0.06 0.08 0.1 0.12 concentration HRP [mg/mL] chemiluminescence signal [V] assay time 30 minutes  400 ms “incubation time” 400 ms 

NO2 NO2 CHO NO2 P(Ph)3 NO2 purple Br- 2-nitrobenzy ltriphenyl- phosphonium brom ide p-nitrobenza ldehyde colourless NaOMe NO2 Me OH colourless P(Ph)3 + + Wittig reaction N + O O Cl Cl Cl Cl O O Cl Cl Cl N Enamine Chloranil blue 2,3,5-trichlor-6-(2-piperidin -1-yl)-[1,4]- benzoquinone Synthesis of a substituted aminovinyl-p-quinone SYNTHESIS 

N+ H H R1 R2 Cl- H H O MeOH N+ Cl- H2 O N R3 R4 R2 R1 C R1 N R2 N R4 R3 H2 O R1 N R2 N R4 R3 O R1/R2 = -CH2 (CH2 )3 CH2 - Piperidine hydrochloride + + Piperidinium cation + R3/R4 = -CH2 (CH2 )4 CH2 - Cyclohexyl isocyanide Nitrilium intermediate -Dialkylacetamide Formaldehyde N-Cyclohexyl-2-piperidin-1-yl-acetamide (1) (2) (3) (4) (5) (6) Multicomponent Chemistries: The Ugi Reaction 0oC 

Inlet capillaries Syringes Rheodyne injection valve Injection loop outlet capillary Micromixer chip / PTFE interface 

Micromixer TOF-MS Injected plug (MeOH) Continuous infusion (MeOH) Secondary amine hydrochloride (1 eq) + Cyclohexylisocyanide (10 eq) Formaldehyde (10 eq) System set-up 

Simultaneous Observation of Reactants, Intermediates, Products and By-products 20 mLmin-1 50 nL injection loop Room temperature 

Compound Library Synthesis Continuous-Flow Dynamic control On-line analysis Real-time identification Real-time optimisation Solution-phase synthesis Well-defined On-line purification... Serial or parallel Solid-Phase Off-line analysis Workup process Off-line identification Off-line optimisation Solid-supported synthesis Physical Handling Ease of purification Parallel (split/pool) Batch Microfluidic Influence of support on chemistry Additional steps required No support No additional steps 

Acknowledgment Coworkers and Ph.D. students Jan Eijkel Chao-Xuan Zhang Giles Sanders Michael Mitchell Fiona Bessoth Omar Naji Darwin Reyes postdocs Arun Arora Yien Kwok Gareth Jenkins Silvia Valussi Nicole Pamme Oliver Hofmann Paul Monaghan Melanie Fennah Valerie Spikmans Nils Goedeke Dimitrios Iossifidis Pierre-Alain Auroux 

FUNDING INSTRUMENTATION SmithKline Beecham (UK) Zeneca (UK) BBSRC, UK EPSRC, UK European Commission, B Schlumberger, UK Casect, UK Agilent, D Forensic Lab, UK Asahi Kasei, Japan Lab of the Government Chemist, UK CSEM, Switzerland Amersham Pharmacia, UK Kodak, UK Glaxo Wellcome, UK Glaxo-Wellcome Heidelberg Instruments Hybaid MICROFABRICATION Alberta Microelectronics Centre, Canada Caliper Technologies, California MESA, University of Twente, The Netherlands CSEM, Switzerland