White Paper While batch granulation techniques are most commonly used in production of pharmaceutical granules, continuous manufacturing offers several advantages including reduced costs, increased flexibility in terms of production volumes and increased consistency in product quality Among the continuous wet granulation methods, twin screw wet granulation (TSWG) is well suited for pharmaceutical processes due to its process stability, flexible scale up, short residence time, and controlled thro.
White Paper Mannitol polymorph transformation in continuous twin-screw granulation Authors: Leo Ohrem, Gudrun Birk and Björn Michel Wet granulation is commonly used during the solid dosage drug manufacturing process to improve powder properties including size, flowability, dissolution rate, bulk density, compressibility and API uniformity Mannitol has become a preferred filler/binder for wet granulation due to its low hygroscopicity, chemical inertness and advantageous tableting behavior including compactibility.1 The wet granulation process generally includes blending, wetting, wet mass stage, drying and sizing Several types of process equipment can be used in batch operation mode: • Low-shear processes use very simple mixing equipment This approach can take a considerable amount of time to achieve a uniformly mixed state • High-shear processes use equipment that mixes the powder and liquid at a very fast rate using high shear forces, and thus accelerate the manufacturing process In continuous operation only: • Twin-screw granulation can continuously manufacture wet granulate powders at lower liquid concentrations and with improved product consistency In both batch operation mode and continuous mode can be used: • Fluid bed granulation is a multiple-step process in which the powders are granulated and dried in the same vessel This approach allows close control of the granulation process It will usually operate under more dry conditions The Life Science business of Merck operates as MilliporeSigma in the U.S and Canada While batch granulation techniques are most commonly used in production of pharmaceutical granules, continuous manufacturing offers several advantages including reduced costs, increased flexibility in terms of production volumes and increased consistency in product quality Among the continuous wet granulation methods, twin-screw wet granulation (TSWG) is well-suited for pharmaceutical processes due to its process stability, flexible scale up, short residence time, and controlled throughput.2 Given the recognized benefits of mannitol in highshear wet granulation, a batch process, its use as an excipient in continuous manufacturing is of growing interest This white paper evaluates the use of Parteck® Delta M excipient, a delta (δ) polymorphic form of D(-) mannitol, in continuous TSWG and investigates whether the polymorphic transition of the δ form to β-mannitol takes place also in continuous mode, enabling delivery of the desired performance characteristics Advantages of Delta Mannitol in Wet Granulation During wet granulation, Parteck® Delta M undergoes a transformation to the β polymorph, resulting in changes in morphology and an increase in surface area (Figure 1) These changes are the major contributing factors for improved compaction behavior during the tableting process.3 Use of δ-mannitol, which undergoes this polymorphic transition, results in superior tabletability compared to granules produced with β-mannitol as a starting material.4 The differences in performance of β- and δ-mannitol in a standard wet granulation batch process and the effect of the polymorphic transition on compaction, dissolution and disintegration properties were assessed in a separate publication.5 Delta mannitol before wet granulation Beta mannitol after wet granulation μm μm Figure SEM images showing the transformation of Parteck® Delta M excipient, a δ crystal of D(-) mannitol to β-mannitol during wet granulation Changes in morphology lead to a significant increase in surface area Continuous Granulation of Parteck® Delta M excipient In TSWG, a solid powder consisting of the API and excipients is fed into the twin-screw extruder while a pump adds the liquid binder Within the barrel, the material is mixed, kneaded and tempered to a target temperature Several process parameters can be adjusted including the liquid-to-solid (L/S) ratio, screw speed and barrel temperature The goal of this study was to verify the δ-to-βtransformation during TSWG Pure δ-mannitol and a model formulation with δ-mannitol (72% Parteck® Delta M excipient, 24% corn starch and 4% polyvinyl pyrrolidone [PVP], all in % w/w) were granulated separately using the continuous tableting line ConsiGma™ CTL 25 (Gea Group, Duesseldorf, Germany) Deionized water was used as the granulation liquid in all cases Granulation trials were performed by variation of three critical process parameters (CPPs): L/S ratio, screw speed and barrel temperature Wet granules were sampled after the TSWG and tray dried for analysis via FT-Raman spectroscopy The wet granules were analyzed for δ-to-β mannitol transformation to understand the effect of drying The formulation pre-blend was prepared in a drum hoop mixer using portions of 30 kg, processed at 30 rpm for 30 minutes The materials were loaded in a layering sequence of ½ mannitol, ½ corn starch, ½ PVP, ½ mannitol, ½ PVP, ½ corn starch into a 100 L metal drum with an internal baffle To investigate the granulation design space, TSWG trials were performed by variation of the process settings screw speed, barrel temperature and L/S ratio Nineteen experiments were performed in total, ten with pure δ-mannitol (L/S ratio: 0.08–0.16, barrel temperature: 30–40 °C, screw speed: 400–800 rpm) and nine with the model formulation (L/S ratio: 0.045–0.10, barrel temperature: 30–40 °C, screw speed: 400–800 rpm) For all trials, pre-blend mass flow rate was set constant at 20 kg/h, fed into the first inlet port of the temperature-controlled granulator barrel by a KT20 twin-screw loss-in-weight feeder (Coperion, Stuttgart, Germany) using coarse concave screws The same granulator screw configuration (1K/6 – 2×1T – 2×1.5T – 4×2T – 6K/4 60° – 1×1.5T – 6K/4 60° – 1×1.5T – 2K/6 60°) with twelve kneading elements was used for all experiments The granulation liquid was added to the TSWG by a liquid pump, dosing the liquid through two ports onto the two screws, just before the first kneading section of the screw configuration Tables and show the process settings for pure δmannitol trials and the model formulation, respectively Slightly different process settings were applied for pure mannitol and formulation granulation runs, based on process stability related observations during the first set point For each experimental point, the TSWG was operated for a total of minutes Wet granules were sampled during the first minute and another sample was collected at the last minute for Raman analysis The latter sample was tray-dried (at 40 °C in Petri dishes, in a drying oven) Approximately 333 g of granules were collected in a plastic beaker for each sample and closed with a sealing film to preserve the relative humidity No Screw speed [rpm] Barrel temp [°C] L/S 800 30 0.08 800 30 0.16 800 30 0.14 800 30 0.12 600 30 0.08 600 30 0.12 600 30 0.14 600 30 0.16 600 40 0.16 10 800 40 0.08 Table Process settings for TSWG trials with pure δ-mannitol Screw speed [rpm] Barrel temp [°C] L/S 800 30 0.080 800 30 0.060 800 30 0.045 800 30 0.100 400 30 0.045 400 30 0.060 500 30 0.080 500 30 0.100 800 40 0.045 Table Process settings for TSWG trials with the model formulation Polymorphic Transformation of Parteck® Delta M excipient The analysis of the polymorphic fractions before and after the granulation process revealed that higher polymorphic transformation occurred with greater L/S ratios, for example, higher β-mannitol content was found in both wet and dry granules (Figure 2) It was found that above a L/S of 0.08, further addition of liquid will not result in significantly higher transformation This observed limit indicates that transformation kinetics are potentially a limiting factor in achieving full conversion The results further indicate that additional transformation occurs during the drying process of the granules when such settings are used as to allow a slow moisture removal rate and thus more time for complete polymorphic transformation Although the difference in β-mannitol content is not significant (p value = 0.1153 and p > 0.05 for wet and dry granules, respectively), the results show a possible tendency 35 30 25 20 15 Wet granules Run 10 (800 rpm 40 °C L/S=0.08) Run (600 rpm 40 °C L/S=0.16) Run (600 rpm 30 °C L/S=0.16) Run (600 rpm 30 °C L/S=0.14) Run (600 rpm 30 °C L/S=0.12) Run (600 rpm 30 °C L/S=0.08) Run (800 rpm 30 °C L/S=0.12) Run (800 rpm 30 °C L/S=0.14) Run (800 rpm 30 °C L/S=0.16) Run (800 rpm 30 °C L/S=0.08) Wet granules Dry granules Figure β-mannitol fraction in wet and dry granules prepared from pure δ-mannitol (mean ±SD, n=3) The findings on the polymorphic transformation in pure mannitol granulation confirmed the results published by Vanhoorne et al (2016).4 Specifically, sufficient transformation from δ- to β-mannitol (between 80– 100%), occurred across a comprehensive range of process parameters in TSWG Run (800 rpm 40 °C L/S=0.045) Run (500 rpm 30 °C L/S=0.1) Run (500 rpm 30 °C L/S=0.08) Run (400 rpm 30 °C L/S=0.06) Run (400 rpm 30 °C L/S=0.045) Run (800 rpm 30 °C L/S=0.1) 20 Run (800 rpm 30 °C L/S=0.045) 40 Run (800 rpm 30 °C L/S=0.06) 60 Run (800 rpm 30 °C L/S=0.08) 80 beta [%] In agreement with the observation of insufficient granulation of the mannitol-containing formulation, a significantly reduced amount of δ-to-β transformation was observed (Figure 3) The lowest conversion of < 5% was observed at a low L/S ratio (0.045), high screw speed (800 rpm) and a barrel temperature of approximately 40 °C With decreasing L/S ratio, a decreasing fraction of transformed mannitol was obtained This can be attributed to a lack of water available for the polymorphic transformation, due to the hygroscopicity of the other components With regard to the effect of other process parameters, no clear trend was observed While a comparison of runs (800 rpm) and (500 rpm), as well as of runs (800 rpm) and (400 rpm), suggests a positive effect of screw speed on δ-to-β transformation, this effect could not be observed in run (800 rpm) vs run (500 rpm) and run (800 rpm) vs run (400 rpm) 10 100 The lowest conversion of about 80% was observed at a low L/S ratio (0.08), in combination with high screw speed (900 and 800 rpm respectively) and a barrel temperature of approximately 30 °C This illustrates that the L/S ratio has a positive correlation and screw speed has a negative correlation with the degree of polymorphic mannitol transformation Additionally, the granule drying process was observed to potentially affect the degree of polymorphic conversion beta [%] No Dry granules Figure β-mannitol fraction in wet and dry granules prepared from a formulation containing δ-mannitol (mean ±SD, n=3) In terms of statistical significance between the β fraction values of wet and dry granules (p value = 0.1866 for wet vs dry granules with p > 0.05, indicating insignificant difference in the β content), no significant difference was detected However, a clear and reversed trend was observed towards lower β-mannitol fractions in dry granules (as opposed to the findings from pure mannitol granules) However, this was attributed to a time delay between the production and the morphologic characterization by Raman Thus, it is possible that further polymorphic conversion occurred during storage of the wet granules As such, it should be confirmed that the transformation is complete so there is no further conversion during storage In one case, however, (run 9) an opposite trend was observed Nevertheless, an interpretation of the effect of process parameters could be evaluated, since the respective samples processed at the same L/S ratio were analyzed with a similar time delay Overall, this study indicated that the main process parameters influencing the δ-to-β transformation are: • L/S ratio (highest influence) • Screw speed (mid-level influence and positive/ negative correlation depending on formulation) • Barrel temperature (no significant influence) Conclusion Given the clear advantages of continuous granulation in comparison to batch operations, identification of excipients for use in TSWG will enable more widespread adoption As described in this study, Parteck® Delta M excipient undergoes the necessary polymorphic transition from its δ polymorphic form to β-mannitol across a comprehensive range of process parameters in continuous TSWG This transition leads to a significantly increased surface area which leads to improved compaction behavior during tableting processes In addition, Parteck® Delta M excipient, the only commercially available mannitol excipient in δ poly morph crystals, offers the inertness of mannitol with excellent binding properties, enabling formulators to more easily develop challenging formulations by wet granulation Reference 1O hrem, HL, et al Why is mannitol becoming more and more popular as a pharmaceutical excipient in solid dosage forms Pharmaceutical Development and Technology, 2014 19(3): 257–262 2L iu, H, et al Optimization of critical quality attributes in continuous twin-screw wet granulation via design space validated with pilot scale experimental data International Journal of Pharmaceutics, 2017 525(1): 249–263 3Y oshinari, T, et al The improved compaction properties of mannitol after a moisture-induced polymorphic transition International Journal of Pharmaceutics, 2003 258(1–2): 121–131 4V anhoorne, V, et al Improved tabletability after a polymorphic transition of delta-mannitol during twin-screw granulation International Journal of Pharmaceutics, 2016 506(1–2): 13–24 5O hrem, HL The Application of Mannitol In Wet Granulation, Pharma’s Almanac, October 2020 For further information, please contact: hans-leonhard.ohrem@merckgroup.com MerckMillipore.com Merck KGaA, Darmstadt, Germany Frankfurter Strasse 250 64293 Darmstadt Germany We provide information and advice to our customers on application technologies and regulatory matters to the best of our knowledge and ability, but without obligation or liability Existing laws and regulations are to be observed in all cases by our customers This also applies in respect to any rights of third parties Our information and advice not relieve our customers of their own responsibility for checking the suitability of our products for the envisaged purpose Merck, the Vibrant M, SAFC and Parteck are trademarks of Merck KGaA, Darmstadt, Germany and/or its affiliates All other trademarks are the property of their respective owners Detailed information on trademarks is available via publicly accessible resources © 2020 Merck KGaA, Darmstadt, Germany and/or its affiliates All Rights Reserved Lit No MK_WP6830EN 09/2020